Search

Find a vulnerability

Search criteria

    45 vulnerabilities found for praisonaiagents by praison

    FKIE_CVE-2026-40288

    Vulnerability from fkie_nvd - Published: 2026-04-14 04:17 - Updated: 2026-10-26 22:10
    Summary
    PraisonAI is a multi-agent teams system. In versions below 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents, the workflow engine is vulnerable to arbitrary command and code execution through untrusted YAML files. When praisonai workflow run <file.yaml> loads a YAML file with type: job, the JobWorkflowExecutor in job_workflow.py processes steps that support run: (shell commands via subprocess.run()), script: (inline Python via exec()), and python: (arbitrary Python script execution)—all without any validation, sandboxing, or user confirmation. The affected code paths include action_run() in workflow.py and _exec_shell(), _exec_inline_python(), and _exec_python_script() in job_workflow.py. An attacker who can supply or influence a workflow YAML file (particularly in CI pipelines, shared repositories, or multi-tenant deployment environments) can achieve full arbitrary command execution on the host system, compromising the machine and any accessible data or credentials. This issue has been fixed in versions 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents.
    Impacted products
    Vendor Product Version
    praison praisonai *
    praison praisonaiagents *

    {
      "affected": [
        {
          "affectedData": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 4.5.139"
                }
              ]
            },
            {
              "product": "praisonaiagents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.140"
                }
              ]
            }
          ],
          "source": "security-advisories@github.com"
        }
      ],
      "configurations": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:a:praison:praisonai:*:*:*:*:*:*:*:*",
                  "matchCriteriaId": "E5F6A0D4-AFE6-4DDA-A1C0-E74C7DD46E00",
                  "versionEndExcluding": "4.5.139",
                  "vulnerable": true
                },
                {
                  "criteria": "cpe:2.3:a:praison:praisonaiagents:*:*:*:*:*:python:*:*",
                  "matchCriteriaId": "37CF4D9C-AD55-414A-9BF2-49A51F03A682",
                  "versionEndExcluding": "1.5.140",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "cveTags": [],
      "descriptions": [
        {
          "lang": "en",
          "value": "PraisonAI is a multi-agent teams system. In versions below 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents, the workflow engine is vulnerable to arbitrary command and code execution through untrusted YAML files. When praisonai workflow run \u003cfile.yaml\u003e loads a YAML file with type: job, the JobWorkflowExecutor in job_workflow.py processes steps that support run: (shell commands via subprocess.run()), script: (inline Python via exec()), and python: (arbitrary Python script execution)\u2014all without any validation, sandboxing, or user confirmation. The affected code paths include action_run() in workflow.py and _exec_shell(), _exec_inline_python(), and _exec_python_script() in job_workflow.py. An attacker who can supply or influence a workflow YAML file (particularly in CI pipelines, shared repositories, or multi-tenant deployment environments) can achieve full arbitrary command execution on the host system, compromising the machine and any accessible data or credentials. This issue has been fixed in versions 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents."
        },
        {
          "lang": "es",
          "value": "PraisonAI es un sistema de equipos multiagente. En versiones anteriores a la 4.5.139 de PraisonAI y la 1.5.140 de praisonaiagents, el motor de flujo de trabajo es vulnerable a la ejecuci\u00f3n arbitraria de comandos y c\u00f3digo a trav\u00e9s de archivos YAML no confiables. Cuando praisonai workflow run \u003cfile.yaml\u003e carga un archivo YAML con type: job, el JobWorkflowExecutor en job_workflow.py procesa pasos que admiten run: (comandos de shell a trav\u00e9s de subprocess.run()), script: (Python en l\u00ednea a trav\u00e9s de exec()), y python: (ejecuci\u00f3n arbitraria de scripts de Python), todo sin ninguna validaci\u00f3n, sandboxing o confirmaci\u00f3n del usuario. Las rutas de c\u00f3digo afectadas incluyen action_run() en workflow.py y _exec_shell(), _exec_inline_python(), y _exec_python_script() en job_workflow.py. Un atacante que puede proporcionar o influir en un archivo YAML de flujo de trabajo (particularmente en pipelines de CI, repositorios compartidos o entornos de despliegue multi-inquilino) puede lograr la ejecuci\u00f3n completa y arbitraria de comandos en el sistema anfitri\u00f3n, comprometiendo la m\u00e1quina y cualquier dato o credencial accesible. Este problema ha sido solucionado en las versiones 4.5.139 de PraisonAI y 1.5.140 de praisonaiagents."
        }
      ],
      "id": "CVE-2026-40288",
      "lastModified": "2026-10-26T22:10:00.163",
      "metrics": {
        "cvssMetricV31": [
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "NETWORK",
              "availabilityImpact": "HIGH",
              "baseScore": 9.8,
              "baseSeverity": "CRITICAL",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "HIGH",
              "privilegesRequired": "NONE",
              "scope": "UNCHANGED",
              "userInteraction": "NONE",
              "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
              "version": "3.1"
            },
            "exploitabilityScore": 3.9,
            "impactScore": 5.9,
            "source": "security-advisories@github.com",
            "type": "Secondary"
          }
        ],
        "ssvcV203": [
          {
            "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
            "ssvcData": {
              "id": "CVE-2026-40288",
              "options": [
                {
                  "exploitation": "poc"
                },
                {
                  "automatable": "yes"
                },
                {
                  "technicalImpact": "total"
                }
              ],
              "role": "CISA Coordinator",
              "timestamp": "2026-04-14T15:56:49.344666Z",
              "version": "2.0.3"
            }
          }
        ]
      },
      "published": "2026-04-14T04:17:12.210",
      "references": [
        {
          "source": "security-advisories@github.com",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-vc46-vw85-3wvm"
        }
      ],
      "sourceIdentifier": "security-advisories@github.com",
      "vulnStatus": "Analyzed",
      "weaknesses": [
        {
          "description": [
            {
              "lang": "en",
              "value": "CWE-78"
            },
            {
              "lang": "en",
              "value": "CWE-94"
            }
          ],
          "source": "security-advisories@github.com",
          "type": "Secondary"
        }
      ]
    }

    FKIE_CVE-2026-40289

    Vulnerability from fkie_nvd - Published: 2026-04-14 04:17 - Updated: 2026-10-26 22:10
    Summary
    PraisonAI is a multi-agent teams system. In versions below 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents, the browser bridge (praisonai browser start) is vulnerable to unauthenticated remote session hijacking due to missing authentication and a bypassable origin check on its /ws WebSocket endpoint. The server binds to 0.0.0.0 by default and only validates the Origin header when one is present, meaning any non-browser client that omits the header is accepted without restriction. An unauthenticated network attacker can connect, send a start_session message, and the server will route it to the first idle browser-extension WebSocket (effectively hijacking that session) and then broadcast all resulting automation actions and outputs back to the attacker. This enables unauthorized remote control of connected browser automation sessions, leakage of sensitive page context and automation results, and misuse of model-backed browser actions in any environment where the bridge is network-reachable. This issue has been fixed in versions 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents.
    Impacted products
    Vendor Product Version
    praison praisonai *
    praison praisonaiagents *

    {
      "affected": [
        {
          "affectedData": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 4.5.139"
                }
              ]
            },
            {
              "product": "praisonaiagents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.140"
                }
              ]
            }
          ],
          "source": "security-advisories@github.com"
        }
      ],
      "configurations": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:a:praison:praisonai:*:*:*:*:*:*:*:*",
                  "matchCriteriaId": "E5F6A0D4-AFE6-4DDA-A1C0-E74C7DD46E00",
                  "versionEndExcluding": "4.5.139",
                  "vulnerable": true
                },
                {
                  "criteria": "cpe:2.3:a:praison:praisonaiagents:*:*:*:*:*:python:*:*",
                  "matchCriteriaId": "37CF4D9C-AD55-414A-9BF2-49A51F03A682",
                  "versionEndExcluding": "1.5.140",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "cveTags": [],
      "descriptions": [
        {
          "lang": "en",
          "value": "PraisonAI is a multi-agent teams system. In versions below 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents, the browser bridge (praisonai browser start) is vulnerable to unauthenticated remote session hijacking due to missing authentication and a bypassable origin check on its /ws WebSocket endpoint. The server binds to 0.0.0.0 by default and only validates the Origin header when one is present, meaning any non-browser client that omits the header is accepted without restriction. An unauthenticated network attacker can connect, send a start_session message, and the server will route it to the first idle browser-extension WebSocket (effectively hijacking that session) and then broadcast all resulting automation actions and outputs back to the attacker. This enables unauthorized remote control of connected browser automation sessions, leakage of sensitive page context and automation results, and misuse of model-backed browser actions in any environment where the bridge is network-reachable. This issue has been fixed in versions 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents."
        },
        {
          "lang": "es",
          "value": "PraisonAI es un sistema de equipos multiagente. En versiones anteriores a la 4.5.139 de PraisonAI y la 1.5.140 de praisonaiagents, el puente del navegador (praisonai browser start) es vulnerable al secuestro remoto de sesi\u00f3n no autenticado debido a la falta de autenticaci\u00f3n y a una verificaci\u00f3n de origen eludible en su endpoint WebSocket /ws. El servidor se enlaza a 0.0.0.0 por defecto y solo valida el encabezado Origin cuando est\u00e1 presente, lo que significa que cualquier cliente que no sea un navegador que omita el encabezado es aceptado sin restricciones. Un atacante de red no autenticado puede conectarse, enviar un mensaje start_session, y el servidor lo enrutar\u00e1 al primer WebSocket de extensi\u00f3n de navegador inactivo (secuestrando efectivamente esa sesi\u00f3n) y luego transmitir\u00e1 todas las acciones y salidas de automatizaci\u00f3n resultantes de vuelta al atacante. Esto permite el control remoto no autorizado de sesiones de automatizaci\u00f3n del navegador conectadas, la fuga de contexto de p\u00e1gina sensible y resultados de automatizaci\u00f3n, y el uso indebido de acciones del navegador respaldadas por modelos en cualquier entorno donde el puente sea accesible por red. Este problema ha sido solucionado en las versiones 4.5.139 de PraisonAI y 1.5.140 de praisonaiagents."
        }
      ],
      "id": "CVE-2026-40289",
      "lastModified": "2026-10-26T22:10:00.163",
      "metrics": {
        "cvssMetricV31": [
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "NETWORK",
              "availabilityImpact": "NONE",
              "baseScore": 9.1,
              "baseSeverity": "CRITICAL",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "HIGH",
              "privilegesRequired": "NONE",
              "scope": "UNCHANGED",
              "userInteraction": "NONE",
              "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N",
              "version": "3.1"
            },
            "exploitabilityScore": 3.9,
            "impactScore": 5.2,
            "source": "security-advisories@github.com",
            "type": "Secondary"
          }
        ],
        "ssvcV203": [
          {
            "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
            "ssvcData": {
              "id": "CVE-2026-40289",
              "options": [
                {
                  "exploitation": "poc"
                },
                {
                  "automatable": "yes"
                },
                {
                  "technicalImpact": "total"
                }
              ],
              "role": "CISA Coordinator",
              "timestamp": "2026-04-14T20:18:27.623888Z",
              "version": "2.0.3"
            }
          }
        ]
      },
      "published": "2026-04-14T04:17:12.710",
      "references": [
        {
          "source": "security-advisories@github.com",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-8x8f-54wf-vv92"
        }
      ],
      "sourceIdentifier": "security-advisories@github.com",
      "vulnStatus": "Analyzed",
      "weaknesses": [
        {
          "description": [
            {
              "lang": "en",
              "value": "CWE-306"
            }
          ],
          "source": "security-advisories@github.com",
          "type": "Secondary"
        }
      ]
    }

    FKIE_CVE-2026-40111

    Vulnerability from fkie_nvd - Published: 2026-04-09 22:16 - Updated: 2026-10-26 22:10
    Summary
    PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, he memory hooks executor in praisonaiagents passes a user-controlled command string directly to subprocess.run() with shell=True at src/praisonai-agents/praisonaiagents/memory/hooks.py. No sanitization is performed and shell metacharacters are interpreted by /bin/sh before the intended command executes. Two independent attack surfaces exist. The first is via pre_run_command and post_run_command hook event types registered through the hooks configuration. The second and more severe surface is the .praisonai/hooks.json lifecycle configuration, where hooks registered for events such as BEFORE_TOOL and AFTER_TOOL fire automatically during agent operation. An agent that gains file-write access through prompt injection can overwrite .praisonai/hooks.json and have its payload execute silently at every subsequent lifecycle event without further user interaction. This vulnerability is fixed in 1.5.128.
    Impacted products
    Vendor Product Version
    praison praisonaiagents *

    {
      "affected": [
        {
          "affectedData": [
            {
              "product": "PraisonAIAgents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.128"
                }
              ]
            }
          ],
          "source": "security-advisories@github.com"
        }
      ],
      "configurations": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:a:praison:praisonaiagents:*:*:*:*:*:*:*:*",
                  "matchCriteriaId": "B1797C27-7EBC-4CEF-90B8-F84E8198632D",
                  "versionEndExcluding": "1.5.128",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "cveTags": [],
      "descriptions": [
        {
          "lang": "en",
          "value": "PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, he memory hooks executor in praisonaiagents passes a user-controlled command string directly to subprocess.run() with shell=True at src/praisonai-agents/praisonaiagents/memory/hooks.py. No sanitization is performed and shell metacharacters are interpreted by /bin/sh before the intended command executes. Two independent attack surfaces exist. The first is via pre_run_command and post_run_command hook event types registered through the hooks configuration. The second and more severe surface is the .praisonai/hooks.json lifecycle configuration, where hooks registered for events such as BEFORE_TOOL and AFTER_TOOL fire automatically during agent operation. An agent that gains file-write access through prompt injection can overwrite .praisonai/hooks.json and have its payload execute silently at every subsequent lifecycle event without further user interaction. This vulnerability is fixed in 1.5.128."
        },
        {
          "lang": "es",
          "value": "PraisonAIAgents es un sistema de equipos multiagente. Antes de la versi\u00f3n 1.5.128, el ejecutor de ganchos de memoria en praisonaiagents pasa una cadena de comando controlada por el usuario directamente a subprocess.run() con shell=True en src/praisonai-agents/praisonaiagents/memory/hooks.py. No se realiza ninguna sanitizaci\u00f3n y los metacaracteres de shell son interpretados por /bin/sh antes de que se ejecute el comando previsto. Dos superficies de ataque independientes existen. La primera es a trav\u00e9s de los tipos de eventos de gancho pre_run_command y post_run_command registrados a trav\u00e9s de la configuraci\u00f3n de ganchos. La segunda y m\u00e1s grave superficie es la configuraci\u00f3n de ciclo de vida .praisonai/hooks.json, donde los ganchos registrados para eventos como BEFORE_TOOL y AFTER_TOOL se disparan autom\u00e1ticamente durante la operaci\u00f3n del agente. Un agente que obtiene acceso de escritura de archivos a trav\u00e9s de inyecci\u00f3n de \u2019prompt\u2019 puede sobrescribir .praisonai/hooks.json y hacer que su carga \u00fatil se ejecute silenciosamente en cada evento de ciclo de vida subsiguiente sin interacci\u00f3n adicional del usuario. Esta vulnerabilidad est\u00e1 corregida en la versi\u00f3n 1.5.128."
        }
      ],
      "id": "CVE-2026-40111",
      "lastModified": "2026-10-26T22:10:00.163",
      "metrics": {
        "cvssMetricV31": [
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "NETWORK",
              "availabilityImpact": "HIGH",
              "baseScore": 8.8,
              "baseSeverity": "HIGH",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "HIGH",
              "privilegesRequired": "NONE",
              "scope": "UNCHANGED",
              "userInteraction": "REQUIRED",
              "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
              "version": "3.1"
            },
            "exploitabilityScore": 2.8,
            "impactScore": 5.9,
            "source": "nvd@nist.gov",
            "type": "Primary"
          }
        ],
        "cvssMetricV40": [
          {
            "cvssData": {
              "Automatable": "NOT_DEFINED",
              "Recovery": "NOT_DEFINED",
              "Safety": "NOT_DEFINED",
              "attackComplexity": "LOW",
              "attackRequirements": "NONE",
              "attackVector": "LOCAL",
              "availabilityRequirement": "NOT_DEFINED",
              "baseScore": 9.3,
              "baseSeverity": "CRITICAL",
              "confidentialityRequirement": "NOT_DEFINED",
              "exploitMaturity": "NOT_DEFINED",
              "integrityRequirement": "NOT_DEFINED",
              "modifiedAttackComplexity": "NOT_DEFINED",
              "modifiedAttackRequirements": "NOT_DEFINED",
              "modifiedAttackVector": "NOT_DEFINED",
              "modifiedPrivilegesRequired": "NOT_DEFINED",
              "modifiedSubAvailabilityImpact": "NOT_DEFINED",
              "modifiedSubConfidentialityImpact": "NOT_DEFINED",
              "modifiedSubIntegrityImpact": "NOT_DEFINED",
              "modifiedUserInteraction": "NOT_DEFINED",
              "modifiedVulnAvailabilityImpact": "NOT_DEFINED",
              "modifiedVulnConfidentialityImpact": "NOT_DEFINED",
              "modifiedVulnIntegrityImpact": "NOT_DEFINED",
              "privilegesRequired": "LOW",
              "providerUrgency": "NOT_DEFINED",
              "subAvailabilityImpact": "HIGH",
              "subConfidentialityImpact": "HIGH",
              "subIntegrityImpact": "HIGH",
              "userInteraction": "NONE",
              "valueDensity": "NOT_DEFINED",
              "vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:L/UI:N/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X",
              "version": "4.0",
              "vulnAvailabilityImpact": "HIGH",
              "vulnConfidentialityImpact": "HIGH",
              "vulnIntegrityImpact": "HIGH",
              "vulnerabilityResponseEffort": "NOT_DEFINED"
            },
            "source": "security-advisories@github.com",
            "type": "Secondary"
          }
        ],
        "ssvcV203": [
          {
            "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
            "ssvcData": {
              "id": "CVE-2026-40111",
              "options": [
                {
                  "exploitation": "poc"
                },
                {
                  "automatable": "no"
                },
                {
                  "technicalImpact": "total"
                }
              ],
              "role": "CISA Coordinator",
              "timestamp": "2026-04-13T15:29:58.666017Z",
              "version": "2.0.3"
            }
          }
        ]
      },
      "published": "2026-04-09T22:16:34.560",
      "references": [
        {
          "source": "security-advisories@github.com",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-v7px-3835-7gjx"
        },
        {
          "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-v7px-3835-7gjx"
        }
      ],
      "sourceIdentifier": "security-advisories@github.com",
      "vulnStatus": "Analyzed",
      "weaknesses": [
        {
          "description": [
            {
              "lang": "en",
              "value": "CWE-78"
            }
          ],
          "source": "security-advisories@github.com",
          "type": "Secondary"
        }
      ]
    }

    FKIE_CVE-2026-40117

    Vulnerability from fkie_nvd - Published: 2026-04-09 22:16 - Updated: 2026-10-26 22:10
    Summary
    PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, read_skill_file() in skill_tools.py allows reading arbitrary files from the filesystem by accepting an unrestricted skill_path parameter. Unlike file_tools.read_file which enforces workspace boundary confinement, and unlike run_skill_script which requires critical-level approval, read_skill_file has neither protection. An agent influenced by prompt injection can exfiltrate sensitive files without triggering any approval prompt. This vulnerability is fixed in 1.5.128.
    Impacted products
    Vendor Product Version
    praison praisonaiagents *

    {
      "affected": [
        {
          "affectedData": [
            {
              "product": "PraisonAIAgents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.128"
                }
              ]
            }
          ],
          "source": "security-advisories@github.com"
        }
      ],
      "configurations": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:a:praison:praisonaiagents:*:*:*:*:*:*:*:*",
                  "matchCriteriaId": "B1797C27-7EBC-4CEF-90B8-F84E8198632D",
                  "versionEndExcluding": "1.5.128",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "cveTags": [],
      "descriptions": [
        {
          "lang": "en",
          "value": "PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, read_skill_file() in skill_tools.py allows reading arbitrary files from the filesystem by accepting an unrestricted skill_path parameter. Unlike file_tools.read_file which enforces workspace boundary confinement, and unlike run_skill_script which requires critical-level approval, read_skill_file has neither protection. An agent influenced by prompt injection can exfiltrate sensitive files without triggering any approval prompt. This vulnerability is fixed in 1.5.128."
        },
        {
          "lang": "es",
          "value": "PraisonAIAgents es un sistema de equipos multiagente. Antes de 1.5.128, read_skill_file() en skill_tools.py permite la lectura de archivos arbitrarios del sistema de archivos al aceptar un par\u00e1metro skill_path sin restricciones. A diferencia de file_tools.read_file, que aplica la contenci\u00f3n de los l\u00edmites del espacio de trabajo, y a diferencia de run_skill_script, que requiere aprobaci\u00f3n de nivel cr\u00edtico, read_skill_file no tiene ninguna de las dos protecciones. Un agente influenciado por inyecci\u00f3n de prompt puede exfiltrar archivos sensibles sin activar ninguna solicitud de aprobaci\u00f3n. Esta vulnerabilidad est\u00e1 corregida en 1.5.128."
        }
      ],
      "id": "CVE-2026-40117",
      "lastModified": "2026-10-26T22:10:00.163",
      "metrics": {
        "cvssMetricV31": [
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "LOCAL",
              "availabilityImpact": "NONE",
              "baseScore": 6.2,
              "baseSeverity": "MEDIUM",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "NONE",
              "privilegesRequired": "NONE",
              "scope": "UNCHANGED",
              "userInteraction": "NONE",
              "vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N",
              "version": "3.1"
            },
            "exploitabilityScore": 2.5,
            "impactScore": 3.6,
            "source": "security-advisories@github.com",
            "type": "Secondary"
          },
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "NETWORK",
              "availabilityImpact": "NONE",
              "baseScore": 7.5,
              "baseSeverity": "HIGH",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "NONE",
              "privilegesRequired": "NONE",
              "scope": "UNCHANGED",
              "userInteraction": "NONE",
              "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N",
              "version": "3.1"
            },
            "exploitabilityScore": 3.9,
            "impactScore": 3.6,
            "source": "nvd@nist.gov",
            "type": "Primary"
          }
        ],
        "ssvcV203": [
          {
            "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
            "ssvcData": {
              "id": "CVE-2026-40117",
              "options": [
                {
                  "exploitation": "poc"
                },
                {
                  "automatable": "no"
                },
                {
                  "technicalImpact": "partial"
                }
              ],
              "role": "CISA Coordinator",
              "timestamp": "2026-04-10T18:14:12.343930Z",
              "version": "2.0.3"
            }
          }
        ]
      },
      "published": "2026-04-09T22:16:35.447",
      "references": [
        {
          "source": "security-advisories@github.com",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-grrg-5cg9-58pf"
        }
      ],
      "sourceIdentifier": "security-advisories@github.com",
      "vulnStatus": "Analyzed",
      "weaknesses": [
        {
          "description": [
            {
              "lang": "en",
              "value": "CWE-862"
            }
          ],
          "source": "security-advisories@github.com",
          "type": "Secondary"
        }
      ]
    }

    FKIE_CVE-2026-40287

    Vulnerability from fkie_nvd - Published: 2026-04-14 04:17 - Updated: 2026-10-26 22:10
    Summary
    PraisonAI is a multi-agent teams system. Versions 4.5.138 and below are vulnerable to arbitrary code execution through automatic, unsanitized import of a tools.py file from the current working directory. Components including call.py (import_tools_from_file()), tool_resolver.py (_load_local_tools()), and CLI tool-loading paths blindly import ./tools.py at startup without any validation, sandboxing, or user confirmation. An attacker who can place a malicious tools.py in the directory where PraisonAI is launched (such as through a shared project, cloned repository, or writable workspace) achieves immediate arbitrary Python code execution in the host environment. This compromises the full PraisonAI process, the host system, and any connected data or credentials. This issue has been fixed in version 4.5.139.
    Impacted products
    Vendor Product Version
    praison praisonai *
    praison praisonaiagents *

    {
      "affected": [
        {
          "affectedData": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 4.5.139"
                }
              ]
            },
            {
              "product": "praisonaiagents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.140"
                }
              ]
            }
          ],
          "source": "security-advisories@github.com"
        }
      ],
      "configurations": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:a:praison:praisonai:*:*:*:*:*:*:*:*",
                  "matchCriteriaId": "E5F6A0D4-AFE6-4DDA-A1C0-E74C7DD46E00",
                  "versionEndExcluding": "4.5.139",
                  "vulnerable": true
                },
                {
                  "criteria": "cpe:2.3:a:praison:praisonaiagents:*:*:*:*:*:python:*:*",
                  "matchCriteriaId": "37CF4D9C-AD55-414A-9BF2-49A51F03A682",
                  "versionEndExcluding": "1.5.140",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "cveTags": [],
      "descriptions": [
        {
          "lang": "en",
          "value": "PraisonAI is a multi-agent teams system. Versions 4.5.138 and below are vulnerable to arbitrary code execution through automatic, unsanitized import of a tools.py file from the current working directory. Components including call.py (import_tools_from_file()), tool_resolver.py (_load_local_tools()), and CLI tool-loading paths blindly import ./tools.py at startup without any validation, sandboxing, or user confirmation. An attacker who can place a malicious tools.py in the directory where PraisonAI is launched (such as through a shared project, cloned repository, or writable workspace) achieves immediate arbitrary Python code execution in the host environment. This compromises the full PraisonAI process, the host system, and any connected data or credentials. This issue has been fixed in version 4.5.139."
        },
        {
          "lang": "es",
          "value": "PraisonAI es un sistema de equipos multiagente. Las versiones 4.5.138 e inferiores son vulnerables a la ejecuci\u00f3n de c\u00f3digo arbitrario mediante la importaci\u00f3n autom\u00e1tica y no saneada de un archivo tools.py desde el directorio de trabajo actual. Componentes que incluyen call.py (import_tools_from_file()), tool_resolver.py (_load_local_tools()) y las rutas de carga de herramientas de la CLI importan ciegamente ./tools.py al inicio sin ninguna validaci\u00f3n, sandboxing o confirmaci\u00f3n del usuario. Un atacante que puede colocar un archivo tools.py malicioso en el directorio donde se inicia PraisonAI (como a trav\u00e9s de un proyecto compartido, un repositorio clonado o un espacio de trabajo escribible) logra la ejecuci\u00f3n inmediata de c\u00f3digo Python arbitrario en el entorno del host. Esto compromete el proceso completo de PraisonAI, el sistema host y cualquier dato o credencial conectado. Este problema ha sido solucionado en la versi\u00f3n 4.5.139."
        }
      ],
      "id": "CVE-2026-40287",
      "lastModified": "2026-10-26T22:10:00.163",
      "metrics": {
        "cvssMetricV31": [
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "LOCAL",
              "availabilityImpact": "HIGH",
              "baseScore": 8.4,
              "baseSeverity": "HIGH",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "HIGH",
              "privilegesRequired": "NONE",
              "scope": "UNCHANGED",
              "userInteraction": "NONE",
              "vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
              "version": "3.1"
            },
            "exploitabilityScore": 2.5,
            "impactScore": 5.9,
            "source": "security-advisories@github.com",
            "type": "Secondary"
          }
        ],
        "ssvcV203": [
          {
            "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
            "ssvcData": {
              "id": "CVE-2026-40287",
              "options": [
                {
                  "exploitation": "poc"
                },
                {
                  "automatable": "no"
                },
                {
                  "technicalImpact": "total"
                }
              ],
              "role": "CISA Coordinator",
              "timestamp": "2026-04-14T13:23:23.221160Z",
              "version": "2.0.3"
            }
          }
        ]
      },
      "published": "2026-04-14T04:17:11.977",
      "references": [
        {
          "source": "security-advisories@github.com",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-g985-wjh9-qxxc"
        },
        {
          "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-g985-wjh9-qxxc"
        }
      ],
      "sourceIdentifier": "security-advisories@github.com",
      "vulnStatus": "Analyzed",
      "weaknesses": [
        {
          "description": [
            {
              "lang": "en",
              "value": "CWE-94"
            },
            {
              "lang": "en",
              "value": "CWE-426"
            }
          ],
          "source": "security-advisories@github.com",
          "type": "Secondary"
        }
      ]
    }

    FKIE_CVE-2026-40152

    Vulnerability from fkie_nvd - Published: 2026-04-09 22:16 - Updated: 2026-10-26 22:10
    Summary
    PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, he list_files() tool in FileTools validates the directory parameter against workspace boundaries via _validate_path(), but passes the pattern parameter directly to Path.glob() without any validation. Since Python's Path.glob() supports .. path segments, an attacker can use relative path traversal in the glob pattern to enumerate arbitrary files outside the workspace, obtaining file metadata (existence, name, size, timestamps) for any path on the filesystem. This vulnerability is fixed in 1.5.128.
    Impacted products
    Vendor Product Version
    praison praisonaiagents *

    {
      "affected": [
        {
          "affectedData": [
            {
              "product": "PraisonAIAgents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.128"
                }
              ]
            }
          ],
          "source": "security-advisories@github.com"
        }
      ],
      "configurations": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:a:praison:praisonaiagents:*:*:*:*:*:*:*:*",
                  "matchCriteriaId": "B1797C27-7EBC-4CEF-90B8-F84E8198632D",
                  "versionEndExcluding": "1.5.128",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "cveTags": [],
      "descriptions": [
        {
          "lang": "en",
          "value": "PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, he list_files() tool in FileTools validates the directory parameter against workspace boundaries via _validate_path(), but passes the pattern parameter directly to Path.glob() without any validation. Since Python\u0027s Path.glob() supports .. path segments, an attacker can use relative path traversal in the glob pattern to enumerate arbitrary files outside the workspace, obtaining file metadata (existence, name, size, timestamps) for any path on the filesystem. This vulnerability is fixed in 1.5.128."
        },
        {
          "lang": "es",
          "value": "PraisonAIAgents es un sistema de equipos multiagente. Antes de la versi\u00f3n 1.5.128, la herramienta list_files() en FileTools valida el par\u00e1metro directory contra los l\u00edmites del espacio de trabajo a trav\u00e9s de _validate_path(), pero pasa el par\u00e1metro pattern directamente a Path.glob() sin ninguna validaci\u00f3n. Dado que Path.glob() de Python admite segmentos de ruta .., un atacante puede usar un salto de ruta relativo en el patr\u00f3n glob para enumerar archivos arbitrarios fuera del espacio de trabajo, obteniendo metadatos de archivos (existencia, nombre, tama\u00f1o, marcas de tiempo) para cualquier ruta en el sistema de archivos. Esta vulnerabilidad se corrige en la versi\u00f3n 1.5.128."
        }
      ],
      "id": "CVE-2026-40152",
      "lastModified": "2026-10-26T22:10:00.163",
      "metrics": {
        "cvssMetricV31": [
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "NETWORK",
              "availabilityImpact": "NONE",
              "baseScore": 5.3,
              "baseSeverity": "MEDIUM",
              "confidentialityImpact": "LOW",
              "integrityImpact": "NONE",
              "privilegesRequired": "NONE",
              "scope": "UNCHANGED",
              "userInteraction": "NONE",
              "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N",
              "version": "3.1"
            },
            "exploitabilityScore": 3.9,
            "impactScore": 1.4,
            "source": "security-advisories@github.com",
            "type": "Secondary"
          }
        ],
        "ssvcV203": [
          {
            "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
            "ssvcData": {
              "id": "CVE-2026-40152",
              "options": [
                {
                  "exploitation": "poc"
                },
                {
                  "automatable": "yes"
                },
                {
                  "technicalImpact": "partial"
                }
              ],
              "role": "CISA Coordinator",
              "timestamp": "2026-04-13T20:40:26.767077Z",
              "version": "2.0.3"
            }
          }
        ]
      },
      "published": "2026-04-09T22:16:36.193",
      "references": [
        {
          "source": "security-advisories@github.com",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-7j2f-xc8p-fjmq"
        }
      ],
      "sourceIdentifier": "security-advisories@github.com",
      "vulnStatus": "Analyzed",
      "weaknesses": [
        {
          "description": [
            {
              "lang": "en",
              "value": "CWE-22"
            }
          ],
          "source": "security-advisories@github.com",
          "type": "Secondary"
        }
      ]
    }

    FKIE_CVE-2026-40160

    Vulnerability from fkie_nvd - Published: 2026-04-10 17:17 - Updated: 2026-10-26 21:10
    Summary
    PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, web_crawl's httpx fallback path passes user-supplied URLs directly to httpx.AsyncClient.get() with follow_redirects=True and no host validation. An LLM agent tricked into crawling an internal URL can reach cloud metadata endpoints (169.254.169.254), internal services, and localhost. The response content is returned to the agent and may appear in output visible to the attacker. This fallback is the default crawl path on a fresh PraisonAI installation (no Tavily key, no Crawl4AI installed). This vulnerability is fixed in 1.5.128.
    Impacted products
    Vendor Product Version
    praison praisonaiagents *

    {
      "affected": [
        {
          "affectedData": [
            {
              "product": "PraisonAIAgents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.128"
                }
              ]
            }
          ],
          "source": "security-advisories@github.com"
        }
      ],
      "configurations": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:a:praison:praisonaiagents:*:*:*:*:*:*:*:*",
                  "matchCriteriaId": "B1797C27-7EBC-4CEF-90B8-F84E8198632D",
                  "versionEndExcluding": "1.5.128",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "cveTags": [],
      "descriptions": [
        {
          "lang": "en",
          "value": "PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, web_crawl\u0027s httpx fallback path passes user-supplied URLs directly to httpx.AsyncClient.get() with follow_redirects=True and no host validation. An LLM agent tricked into crawling an internal URL can reach cloud metadata endpoints (169.254.169.254), internal services, and localhost. The response content is returned to the agent and may appear in output visible to the attacker. This fallback is the default crawl path on a fresh PraisonAI installation (no Tavily key, no Crawl4AI installed). This vulnerability is fixed in 1.5.128."
        },
        {
          "lang": "es",
          "value": "PraisonAIAgents es un sistema de equipos multiagente. Antes de la versi\u00f3n 1.5.128, la ruta de respaldo httpx de web_crawl pasa las URL proporcionadas por el usuario directamente a httpx.AsyncClient.get() con follow_redirects=True y sin validaci\u00f3n de host. Un agente LLM enga\u00f1ado para rastrear una URL interna puede alcanzar puntos finales de metadatos en la nube (169.254.169.254), servicios internos y localhost. El contenido de la respuesta se devuelve al agente y puede aparecer en la salida visible para el atacante. Esta ruta de respaldo es la ruta de rastreo predeterminada en una instalaci\u00f3n nueva de PraisonAI (sin clave de Tavily, sin Crawl4AI instalado). Esta vulnerabilidad est\u00e1 corregida en la versi\u00f3n 1.5.128."
        }
      ],
      "id": "CVE-2026-40160",
      "lastModified": "2026-10-26T21:10:00.170",
      "metrics": {
        "cvssMetricV31": [
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "NETWORK",
              "availabilityImpact": "NONE",
              "baseScore": 6.5,
              "baseSeverity": "MEDIUM",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "NONE",
              "privilegesRequired": "NONE",
              "scope": "UNCHANGED",
              "userInteraction": "REQUIRED",
              "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:N/A:N",
              "version": "3.1"
            },
            "exploitabilityScore": 2.8,
            "impactScore": 3.6,
            "source": "nvd@nist.gov",
            "type": "Primary"
          }
        ],
        "cvssMetricV40": [
          {
            "cvssData": {
              "Automatable": "NOT_DEFINED",
              "Recovery": "NOT_DEFINED",
              "Safety": "NOT_DEFINED",
              "attackComplexity": "LOW",
              "attackRequirements": "PRESENT",
              "attackVector": "NETWORK",
              "availabilityRequirement": "NOT_DEFINED",
              "baseScore": 7.1,
              "baseSeverity": "HIGH",
              "confidentialityRequirement": "NOT_DEFINED",
              "exploitMaturity": "NOT_DEFINED",
              "integrityRequirement": "NOT_DEFINED",
              "modifiedAttackComplexity": "NOT_DEFINED",
              "modifiedAttackRequirements": "NOT_DEFINED",
              "modifiedAttackVector": "NOT_DEFINED",
              "modifiedPrivilegesRequired": "NOT_DEFINED",
              "modifiedSubAvailabilityImpact": "NOT_DEFINED",
              "modifiedSubConfidentialityImpact": "NOT_DEFINED",
              "modifiedSubIntegrityImpact": "NOT_DEFINED",
              "modifiedUserInteraction": "NOT_DEFINED",
              "modifiedVulnAvailabilityImpact": "NOT_DEFINED",
              "modifiedVulnConfidentialityImpact": "NOT_DEFINED",
              "modifiedVulnIntegrityImpact": "NOT_DEFINED",
              "privilegesRequired": "NONE",
              "providerUrgency": "NOT_DEFINED",
              "subAvailabilityImpact": "NONE",
              "subConfidentialityImpact": "HIGH",
              "subIntegrityImpact": "LOW",
              "userInteraction": "PASSIVE",
              "valueDensity": "NOT_DEFINED",
              "vectorString": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:P/VC:H/VI:N/VA:N/SC:H/SI:L/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X",
              "version": "4.0",
              "vulnAvailabilityImpact": "NONE",
              "vulnConfidentialityImpact": "HIGH",
              "vulnIntegrityImpact": "NONE",
              "vulnerabilityResponseEffort": "NOT_DEFINED"
            },
            "source": "security-advisories@github.com",
            "type": "Secondary"
          }
        ],
        "ssvcV203": [
          {
            "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
            "ssvcData": {
              "id": "CVE-2026-40160",
              "options": [
                {
                  "exploitation": "poc"
                },
                {
                  "automatable": "no"
                },
                {
                  "technicalImpact": "partial"
                }
              ],
              "role": "CISA Coordinator",
              "timestamp": "2026-04-13T15:28:31.603808Z",
              "version": "2.0.3"
            }
          }
        ]
      },
      "published": "2026-04-10T17:17:13.950",
      "references": [
        {
          "source": "security-advisories@github.com",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-qq9r-63f6-v542"
        },
        {
          "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-qq9r-63f6-v542"
        }
      ],
      "sourceIdentifier": "security-advisories@github.com",
      "vulnStatus": "Analyzed",
      "weaknesses": [
        {
          "description": [
            {
              "lang": "en",
              "value": "CWE-918"
            }
          ],
          "source": "security-advisories@github.com",
          "type": "Secondary"
        }
      ]
    }

    FKIE_CVE-2026-40150

    Vulnerability from fkie_nvd - Published: 2026-04-09 22:16 - Updated: 2026-10-26 21:10
    Summary
    PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, the web_crawl() function in praisonaiagents/tools/web_crawl_tools.py accepts arbitrary URLs from AI agents with zero validation. No scheme allowlisting, hostname/IP blocklisting, or private network checks are applied before fetching. This allows an attacker (or prompt injection in crawled content) to force the agent to fetch cloud metadata endpoints, internal services, or local files via file:// URLs. This vulnerability is fixed in 1.5.128.
    Impacted products
    Vendor Product Version
    praison praisonaiagents *

    {
      "affected": [
        {
          "affectedData": [
            {
              "product": "PraisonAIAgents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.128"
                }
              ]
            }
          ],
          "source": "security-advisories@github.com"
        }
      ],
      "configurations": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:a:praison:praisonaiagents:*:*:*:*:*:*:*:*",
                  "matchCriteriaId": "B1797C27-7EBC-4CEF-90B8-F84E8198632D",
                  "versionEndExcluding": "1.5.128",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "cveTags": [],
      "descriptions": [
        {
          "lang": "en",
          "value": "PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, the web_crawl() function in praisonaiagents/tools/web_crawl_tools.py accepts arbitrary URLs from AI agents with zero validation. No scheme allowlisting, hostname/IP blocklisting, or private network checks are applied before fetching. This allows an attacker (or prompt injection in crawled content) to force the agent to fetch cloud metadata endpoints, internal services, or local files via file:// URLs. This vulnerability is fixed in 1.5.128."
        },
        {
          "lang": "es",
          "value": "PraisonAIAgents es un sistema de equipos multiagente. Antes de la versi\u00f3n 1.5.128, la funci\u00f3n web_crawl() en praisonaiagents/tools/web_crawl_tools.py acepta URL arbitrarias de agentes de IA sin ninguna validaci\u00f3n. No se aplica ninguna lista blanca de esquemas, lista negra de nombres de host/IP o comprobaciones de red privada antes de la recuperaci\u00f3n. Esto permite a un atacante (o la inyecci\u00f3n de \u2019prompts\u2019 en el contenido rastreado) forzar al agente a recuperar puntos finales de metadatos en la nube, servicios internos o archivos locales a trav\u00e9s de URL de tipo file: //. Esta vulnerabilidad se corrige en la versi\u00f3n 1.5.128."
        }
      ],
      "id": "CVE-2026-40150",
      "lastModified": "2026-10-26T21:10:00.170",
      "metrics": {
        "cvssMetricV31": [
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "NETWORK",
              "availabilityImpact": "NONE",
              "baseScore": 7.7,
              "baseSeverity": "HIGH",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "NONE",
              "privilegesRequired": "LOW",
              "scope": "CHANGED",
              "userInteraction": "NONE",
              "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:N/A:N",
              "version": "3.1"
            },
            "exploitabilityScore": 3.1,
            "impactScore": 4.0,
            "source": "security-advisories@github.com",
            "type": "Secondary"
          },
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "NETWORK",
              "availabilityImpact": "NONE",
              "baseScore": 6.5,
              "baseSeverity": "MEDIUM",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "NONE",
              "privilegesRequired": "LOW",
              "scope": "UNCHANGED",
              "userInteraction": "NONE",
              "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N",
              "version": "3.1"
            },
            "exploitabilityScore": 2.8,
            "impactScore": 3.6,
            "source": "nvd@nist.gov",
            "type": "Primary"
          }
        ],
        "ssvcV203": [
          {
            "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
            "ssvcData": {
              "id": "CVE-2026-40150",
              "options": [
                {
                  "exploitation": "poc"
                },
                {
                  "automatable": "no"
                },
                {
                  "technicalImpact": "partial"
                }
              ],
              "role": "CISA Coordinator",
              "timestamp": "2026-04-14T14:40:16.617609Z",
              "version": "2.0.3"
            }
          }
        ]
      },
      "published": "2026-04-09T22:16:35.900",
      "references": [
        {
          "source": "security-advisories@github.com",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-8f4v-xfm9-3244"
        },
        {
          "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-8f4v-xfm9-3244"
        }
      ],
      "sourceIdentifier": "security-advisories@github.com",
      "vulnStatus": "Analyzed",
      "weaknesses": [
        {
          "description": [
            {
              "lang": "en",
              "value": "CWE-918"
            }
          ],
          "source": "security-advisories@github.com",
          "type": "Secondary"
        }
      ]
    }

    FKIE_CVE-2026-40153

    Vulnerability from fkie_nvd - Published: 2026-04-09 22:16 - Updated: 2026-10-26 21:10
    Summary
    PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, the execute_command function in shell_tools.py calls os.path.expandvars() on every command argument at line 64, manually re-implementing shell-level environment variable expansion despite using shell=False (line 88) for security. This allows exfiltration of secrets stored in environment variables (database credentials, API keys, cloud access keys). The approval system displays the unexpanded $VAR references to human reviewers, creating a deceptive approval where the displayed command differs from what actually executes. This vulnerability is fixed in 1.5.128.
    Impacted products
    Vendor Product Version
    praison praisonaiagents *

    {
      "affected": [
        {
          "affectedData": [
            {
              "product": "PraisonAIAgents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.128"
                }
              ]
            }
          ],
          "source": "security-advisories@github.com"
        }
      ],
      "configurations": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:a:praison:praisonaiagents:*:*:*:*:*:*:*:*",
                  "matchCriteriaId": "B1797C27-7EBC-4CEF-90B8-F84E8198632D",
                  "versionEndExcluding": "1.5.128",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "cveTags": [],
      "descriptions": [
        {
          "lang": "en",
          "value": "PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, the execute_command function in shell_tools.py calls os.path.expandvars() on every command argument at line 64, manually re-implementing shell-level environment variable expansion despite using shell=False (line 88) for security. This allows exfiltration of secrets stored in environment variables (database credentials, API keys, cloud access keys). The approval system displays the unexpanded $VAR references to human reviewers, creating a deceptive approval where the displayed command differs from what actually executes. This vulnerability is fixed in 1.5.128."
        },
        {
          "lang": "es",
          "value": "PraisonAIAgents es un sistema de equipos multiagente. Antes de la versi\u00f3n 1.5.128, la funci\u00f3n execute_command en shell_tools.py llama a os.path.expandvars() en cada argumento de comando en la l\u00ednea 64, reimplementando manualmente la expansi\u00f3n de variables de entorno a nivel de shell a pesar de usar shell=False (l\u00ednea 88) por seguridad. Esto permite la exfiltraci\u00f3n de secretos almacenados en variables de entorno (credenciales de base de datos, claves de API, claves de acceso a la nube). El sistema de aprobaci\u00f3n muestra las referencias $VAR sin expandir a los revisores humanos, creando una aprobaci\u00f3n enga\u00f1osa donde el comando mostrado difiere de lo que realmente se ejecuta. Esta vulnerabilidad se corrige en la versi\u00f3n 1.5.128."
        }
      ],
      "id": "CVE-2026-40153",
      "lastModified": "2026-10-26T21:10:00.170",
      "metrics": {
        "cvssMetricV31": [
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "NETWORK",
              "availabilityImpact": "NONE",
              "baseScore": 7.4,
              "baseSeverity": "HIGH",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "NONE",
              "privilegesRequired": "NONE",
              "scope": "CHANGED",
              "userInteraction": "REQUIRED",
              "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:N/A:N",
              "version": "3.1"
            },
            "exploitabilityScore": 2.8,
            "impactScore": 4.0,
            "source": "security-advisories@github.com",
            "type": "Secondary"
          },
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "NETWORK",
              "availabilityImpact": "NONE",
              "baseScore": 6.5,
              "baseSeverity": "MEDIUM",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "NONE",
              "privilegesRequired": "NONE",
              "scope": "UNCHANGED",
              "userInteraction": "REQUIRED",
              "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:N/A:N",
              "version": "3.1"
            },
            "exploitabilityScore": 2.8,
            "impactScore": 3.6,
            "source": "nvd@nist.gov",
            "type": "Primary"
          }
        ],
        "ssvcV203": [
          {
            "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
            "ssvcData": {
              "id": "CVE-2026-40153",
              "options": [
                {
                  "exploitation": "poc"
                },
                {
                  "automatable": "no"
                },
                {
                  "technicalImpact": "partial"
                }
              ],
              "role": "CISA Coordinator",
              "timestamp": "2026-04-13T15:28:34.665913Z",
              "version": "2.0.3"
            }
          }
        ]
      },
      "published": "2026-04-09T22:16:36.350",
      "references": [
        {
          "source": "security-advisories@github.com",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-v8g7-9q6v-p3x8"
        },
        {
          "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-v8g7-9q6v-p3x8"
        }
      ],
      "sourceIdentifier": "security-advisories@github.com",
      "vulnStatus": "Analyzed",
      "weaknesses": [
        {
          "description": [
            {
              "lang": "en",
              "value": "CWE-526"
            }
          ],
          "source": "security-advisories@github.com",
          "type": "Secondary"
        }
      ]
    }

    FKIE_CVE-2026-41496

    Vulnerability from fkie_nvd - Published: 2026-05-08 14:16 - Updated: 2026-10-26 19:10
    Summary
    PraisonAI is a multi-agent teams system. Prior to praisonai version 4.6.9 and praisonaiagents version 1.6.9, the fix for CVE-2026-40315 added input validation to SQLiteConversationStore only. Nine sibling backends — MySQL, PostgreSQL, async SQLite/MySQL/PostgreSQL, Turso, SingleStore, Supabase, SurrealDB — pass table_prefix straight into f-string SQL. Same root cause, same code pattern, same exploitation. 52 unvalidated injection points across the codebase. postgres.py additionally accepts an unvalidated schema parameter used directly in DDL. This issue has been patched in praisonai version 4.6.9 and praisonaiagents version 1.6.9.
    Impacted products
    Vendor Product Version
    praison praisonai *
    praison praisonaiagents *

    {
      "affected": [
        {
          "affectedData": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "praisonaiagents \u003c 1.6.9"
                },
                {
                  "status": "affected",
                  "version": "praisonai \u003c 4.6.9"
                }
              ]
            }
          ],
          "source": "security-advisories@github.com"
        }
      ],
      "configurations": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:a:praison:praisonai:*:*:*:*:*:*:*:*",
                  "matchCriteriaId": "926F2340-8373-45E8-8F68-EDB9E771A52E",
                  "versionEndExcluding": "4.6.9",
                  "vulnerable": true
                },
                {
                  "criteria": "cpe:2.3:a:praison:praisonaiagents:*:*:*:*:*:python:*:*",
                  "matchCriteriaId": "CE98749C-A5F2-4D34-9CBA-87CA20237D1E",
                  "versionEndExcluding": "1.6.9",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "cveTags": [],
      "descriptions": [
        {
          "lang": "en",
          "value": "PraisonAI is a multi-agent teams system. Prior to praisonai version 4.6.9 and praisonaiagents version 1.6.9, the fix for CVE-2026-40315 added input validation to SQLiteConversationStore only. Nine sibling backends \u2014 MySQL, PostgreSQL, async SQLite/MySQL/PostgreSQL, Turso, SingleStore, Supabase, SurrealDB \u2014 pass table_prefix straight into f-string SQL. Same root cause, same code pattern, same exploitation. 52 unvalidated injection points across the codebase. postgres.py additionally accepts an unvalidated schema parameter used directly in DDL. This issue has been patched in praisonai version 4.6.9 and praisonaiagents version 1.6.9."
        },
        {
          "lang": "es",
          "value": "PraisonAI es un sistema de equipos multiagente. Antes de la versi\u00f3n 4.6.9 de praisonai y la versi\u00f3n 1.6.9 de praisonaiagents, la correcci\u00f3n para CVE-2026-40315 a\u00f1adi\u00f3 validaci\u00f3n de entrada solo a SQLiteConversationStore. Nueve backends hermanos - MySQL, PostgreSQL, SQLite/MySQL/PostgreSQL as\u00edncrono, Turso, SingleStore, Supabase, SurrealDB - pasan table_prefix directamente a SQL de f-string. Misma causa ra\u00edz, mismo patr\u00f3n de c\u00f3digo, misma explotaci\u00f3n. 52 puntos de inyecci\u00f3n no validados en toda la base de c\u00f3digo. postgres.py acepta adicionalmente un par\u00e1metro de esquema no validado utilizado directamente en DDL. Este problema ha sido parcheado en la versi\u00f3n 4.6.9 de praisonai y la versi\u00f3n 1.6.9 de praisonaiagents."
        }
      ],
      "id": "CVE-2026-41496",
      "lastModified": "2026-10-26T19:10:00.163",
      "metrics": {
        "cvssMetricV31": [
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "NETWORK",
              "availabilityImpact": "NONE",
              "baseScore": 8.1,
              "baseSeverity": "HIGH",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "HIGH",
              "privilegesRequired": "LOW",
              "scope": "UNCHANGED",
              "userInteraction": "NONE",
              "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:N",
              "version": "3.1"
            },
            "exploitabilityScore": 2.8,
            "impactScore": 5.2,
            "source": "security-advisories@github.com",
            "type": "Secondary"
          }
        ],
        "ssvcV203": [
          {
            "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
            "ssvcData": {
              "id": "CVE-2026-41496",
              "options": [
                {
                  "exploitation": "poc"
                },
                {
                  "automatable": "no"
                },
                {
                  "technicalImpact": "total"
                }
              ],
              "role": "CISA Coordinator",
              "timestamp": "2026-05-08T23:17:23.691346Z",
              "version": "2.0.3"
            }
          }
        ]
      },
      "published": "2026-05-08T14:16:33.693",
      "references": [
        {
          "source": "security-advisories@github.com",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-rg3h-x3jw-7jm5"
        },
        {
          "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-rg3h-x3jw-7jm5"
        }
      ],
      "sourceIdentifier": "security-advisories@github.com",
      "vulnStatus": "Modified",
      "weaknesses": [
        {
          "description": [
            {
              "lang": "en",
              "value": "CWE-89"
            }
          ],
          "source": "security-advisories@github.com",
          "type": "Secondary"
        }
      ]
    }

    FKIE_CVE-2026-44339

    Vulnerability from fkie_nvd - Published: 2026-05-08 14:16 - Updated: 2026-10-26 18:10
    Summary
    PraisonAI is a multi-agent teams system. Prior to praisonai version 4.6.37 and praisonaiagents version 1.6.37, praisonaiagents resolves unresolved tool names against module globals and __main__ after it fails to match the declared tool list and the registry. With the default agent configuration, _perm_allow is None, so undeclared non-dangerous tool names are not rejected by the permission gate. An attacker who can influence tool-call names can therefore invoke unintended application callables that were never declared as tools. This issue has been patched in praisonai version 4.6.37 and praisonaiagents version 1.6.37.
    Impacted products
    Vendor Product Version
    praison praisonai *
    praison praisonaiagents *

    {
      "affected": [
        {
          "affectedData": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "praisonaiagents \u003c 1.6.37"
                },
                {
                  "status": "affected",
                  "version": "praisonai \u003c 4.6.37"
                }
              ]
            }
          ],
          "source": "security-advisories@github.com"
        }
      ],
      "configurations": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:a:praison:praisonai:*:*:*:*:*:*:*:*",
                  "matchCriteriaId": "A4E257B3-EAC7-4FEF-B4DD-E262F1922CDF",
                  "versionEndExcluding": "4.6.37",
                  "vulnerable": true
                },
                {
                  "criteria": "cpe:2.3:a:praison:praisonaiagents:*:*:*:*:*:python:*:*",
                  "matchCriteriaId": "7FC5B4FE-995C-40C9-92E3-4C30D9445F89",
                  "versionEndExcluding": "1.6.37",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "cveTags": [],
      "descriptions": [
        {
          "lang": "en",
          "value": "PraisonAI is a multi-agent teams system. Prior to praisonai version 4.6.37 and praisonaiagents version 1.6.37, praisonaiagents resolves unresolved tool names against module globals and __main__ after it fails to match the declared tool list and the registry. With the default agent configuration, _perm_allow is None, so undeclared non-dangerous tool names are not rejected by the permission gate. An attacker who can influence tool-call names can therefore invoke unintended application callables that were never declared as tools. This issue has been patched in praisonai version 4.6.37 and praisonaiagents version 1.6.37."
        },
        {
          "lang": "es",
          "value": "PraisonAI es un sistema de equipos multiagente. Antes de la versi\u00f3n 4.6.37 de praisonai y la versi\u00f3n 1.6.37 de praisonaiagents, praisonaiagents resuelve los nombres de herramientas no resueltos contra las variables globales del m\u00f3dulo y __main__ despu\u00e9s de que no logra coincidir con la lista de herramientas declaradas y el registro. Con la configuraci\u00f3n de agente predeterminada, _perm_allow es None, por lo tanto, los nombres de herramientas no peligrosas no declaradas no son rechazados por la puerta de permisos. Un atacante que puede influir en los nombres de las llamadas a herramientas puede, por lo tanto, invocar elementos invocables de la aplicaci\u00f3n no intencionados que nunca fueron declarados como herramientas. Este problema ha sido parcheado en la versi\u00f3n 4.6.37 de praisonai y la versi\u00f3n 1.6.37 de praisonaiagents."
        }
      ],
      "id": "CVE-2026-44339",
      "lastModified": "2026-10-26T18:10:00.173",
      "metrics": {
        "cvssMetricV31": [
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "NETWORK",
              "availabilityImpact": "LOW",
              "baseScore": 8.6,
              "baseSeverity": "HIGH",
              "confidentialityImpact": "LOW",
              "integrityImpact": "HIGH",
              "privilegesRequired": "NONE",
              "scope": "UNCHANGED",
              "userInteraction": "NONE",
              "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:H/A:L",
              "version": "3.1"
            },
            "exploitabilityScore": 3.9,
            "impactScore": 4.7,
            "source": "security-advisories@github.com",
            "type": "Secondary"
          }
        ],
        "ssvcV203": [
          {
            "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
            "ssvcData": {
              "id": "CVE-2026-44339",
              "options": [
                {
                  "exploitation": "poc"
                },
                {
                  "automatable": "yes"
                },
                {
                  "technicalImpact": "partial"
                }
              ],
              "role": "CISA Coordinator",
              "timestamp": "2026-05-08T17:03:56.082162Z",
              "version": "2.0.3"
            }
          }
        ]
      },
      "published": "2026-05-08T14:16:46.887",
      "references": [
        {
          "source": "security-advisories@github.com",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-gmjg-hv98-qggq"
        },
        {
          "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-gmjg-hv98-qggq"
        }
      ],
      "sourceIdentifier": "security-advisories@github.com",
      "vulnStatus": "Modified",
      "weaknesses": [
        {
          "description": [
            {
              "lang": "en",
              "value": "CWE-470"
            }
          ],
          "source": "security-advisories@github.com",
          "type": "Secondary"
        }
      ]
    }

    FKIE_CVE-2026-44335

    Vulnerability from fkie_nvd - Published: 2026-05-08 14:16 - Updated: 2026-10-26 18:10
    Summary
    PraisonAI is a multi-agent teams system. Prior to version 1.6.32, the URL checking logic in PraisonAI has a logical flaw that could be bypassed by attackers, leading to SSRF attacks. This issue has been patched in version 1.6.32.
    Impacted products
    Vendor Product Version
    praison praisonaiagents *

    {
      "affected": [
        {
          "affectedData": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.6.32"
                }
              ]
            }
          ],
          "source": "security-advisories@github.com"
        }
      ],
      "configurations": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:a:praison:praisonaiagents:*:*:*:*:*:python:*:*",
                  "matchCriteriaId": "AB49F79A-F9E9-4E57-8FC7-CAA38FBC171C",
                  "versionEndExcluding": "1.6.32",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "cveTags": [],
      "descriptions": [
        {
          "lang": "en",
          "value": "PraisonAI is a multi-agent teams system. Prior to version 1.6.32, the URL checking logic in PraisonAI has a logical flaw that could be bypassed by attackers, leading to SSRF attacks. This issue has been patched in version 1.6.32."
        },
        {
          "lang": "es",
          "value": "PraisonAI es un sistema de equipos multiagente. Antes de la versi\u00f3n 1.6.32, la l\u00f3gica de verificaci\u00f3n de URL en PraisonAI tiene una falla l\u00f3gica que podr\u00eda ser eludida por atacantes, lo que llevar\u00eda a ataques SSRF. Este problema ha sido parcheado en la versi\u00f3n 1.6.32."
        }
      ],
      "id": "CVE-2026-44335",
      "lastModified": "2026-10-26T18:10:00.173",
      "metrics": {
        "cvssMetricV31": [
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "NETWORK",
              "availabilityImpact": "HIGH",
              "baseScore": 9.8,
              "baseSeverity": "CRITICAL",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "HIGH",
              "privilegesRequired": "NONE",
              "scope": "UNCHANGED",
              "userInteraction": "NONE",
              "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
              "version": "3.1"
            },
            "exploitabilityScore": 3.9,
            "impactScore": 5.9,
            "source": "nvd@nist.gov",
            "type": "Primary"
          }
        ],
        "cvssMetricV40": [
          {
            "cvssData": {
              "Automatable": "NOT_DEFINED",
              "Recovery": "NOT_DEFINED",
              "Safety": "NOT_DEFINED",
              "attackComplexity": "LOW",
              "attackRequirements": "NONE",
              "attackVector": "NETWORK",
              "availabilityRequirement": "NOT_DEFINED",
              "baseScore": 7.7,
              "baseSeverity": "HIGH",
              "confidentialityRequirement": "NOT_DEFINED",
              "exploitMaturity": "PROOF_OF_CONCEPT",
              "integrityRequirement": "NOT_DEFINED",
              "modifiedAttackComplexity": "NOT_DEFINED",
              "modifiedAttackRequirements": "NOT_DEFINED",
              "modifiedAttackVector": "NOT_DEFINED",
              "modifiedPrivilegesRequired": "NOT_DEFINED",
              "modifiedSubAvailabilityImpact": "NOT_DEFINED",
              "modifiedSubConfidentialityImpact": "NOT_DEFINED",
              "modifiedSubIntegrityImpact": "NOT_DEFINED",
              "modifiedUserInteraction": "NOT_DEFINED",
              "modifiedVulnAvailabilityImpact": "NOT_DEFINED",
              "modifiedVulnConfidentialityImpact": "NOT_DEFINED",
              "modifiedVulnIntegrityImpact": "NOT_DEFINED",
              "privilegesRequired": "NONE",
              "providerUrgency": "NOT_DEFINED",
              "subAvailabilityImpact": "NONE",
              "subConfidentialityImpact": "NONE",
              "subIntegrityImpact": "NONE",
              "userInteraction": "NONE",
              "valueDensity": "NOT_DEFINED",
              "vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:N/SC:N/SI:N/SA:N/E:P/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X",
              "version": "4.0",
              "vulnAvailabilityImpact": "NONE",
              "vulnConfidentialityImpact": "NONE",
              "vulnIntegrityImpact": "HIGH",
              "vulnerabilityResponseEffort": "NOT_DEFINED"
            },
            "source": "security-advisories@github.com",
            "type": "Secondary"
          }
        ],
        "ssvcV203": [
          {
            "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
            "ssvcData": {
              "id": "CVE-2026-44335",
              "options": [
                {
                  "exploitation": "poc"
                },
                {
                  "automatable": "yes"
                },
                {
                  "technicalImpact": "partial"
                }
              ],
              "role": "CISA Coordinator",
              "timestamp": "2026-05-08T14:46:06.544307Z",
              "version": "2.0.3"
            }
          }
        ]
      },
      "published": "2026-05-08T14:16:46.290",
      "references": [
        {
          "source": "security-advisories@github.com",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-q9pw-vmhh-384g"
        },
        {
          "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-q9pw-vmhh-384g"
        }
      ],
      "sourceIdentifier": "security-advisories@github.com",
      "vulnStatus": "Analyzed",
      "weaknesses": [
        {
          "description": [
            {
              "lang": "en",
              "value": "CWE-918"
            }
          ],
          "source": "security-advisories@github.com",
          "type": "Secondary"
        }
      ]
    }

    FKIE_CVE-2026-34938

    Vulnerability from fkie_nvd - Published: 2026-04-03 23:17 - Updated: 2026-07-24 22:10
    Summary
    PraisonAI is a multi-agent teams system. Prior to version 1.5.90, execute_code() in praisonai-agents runs attacker-controlled Python inside a three-layer sandbox that can be fully bypassed by passing a str subclass with an overridden startswith() method to the _safe_getattr wrapper, achieving arbitrary OS command execution on the host. This issue has been patched in version 1.5.90.
    Impacted products
    Vendor Product Version
    praison praisonaiagents *

    {
      "affected": [
        {
          "affectedData": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.90"
                }
              ]
            }
          ],
          "source": "security-advisories@github.com"
        }
      ],
      "configurations": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:a:praison:praisonaiagents:*:*:*:*:*:*:*:*",
                  "matchCriteriaId": "0CECD474-C348-4AD9-B570-42895741C580",
                  "versionEndExcluding": "1.5.90",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "cveTags": [],
      "descriptions": [
        {
          "lang": "en",
          "value": "PraisonAI is a multi-agent teams system. Prior to version 1.5.90, execute_code() in praisonai-agents runs attacker-controlled Python inside a three-layer sandbox that can be fully bypassed by passing a str subclass with an overridden startswith() method to the _safe_getattr wrapper, achieving arbitrary OS command execution on the host. This issue has been patched in version 1.5.90."
        },
        {
          "lang": "es",
          "value": "PraisonAI es un sistema de equipos multiagente. Antes de la versi\u00f3n 1.5.90, execute_code() en praisonai-agents ejecuta Python controlado por el atacante dentro de una sandbox de tres capas que puede ser completamente eludida pasando una subclase de str con un m\u00e9todo startswith() sobrescrito al wrapper _safe_getattr, logrando la ejecuci\u00f3n arbitraria de comandos del sistema operativo en el host. Este problema ha sido parcheado en la versi\u00f3n 1.5.90."
        }
      ],
      "id": "CVE-2026-34938",
      "lastModified": "2026-07-24T22:10:00.140",
      "metrics": {
        "cvssMetricV31": [
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "NETWORK",
              "availabilityImpact": "HIGH",
              "baseScore": 10.0,
              "baseSeverity": "CRITICAL",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "HIGH",
              "privilegesRequired": "NONE",
              "scope": "CHANGED",
              "userInteraction": "NONE",
              "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H",
              "version": "3.1"
            },
            "exploitabilityScore": 3.9,
            "impactScore": 6.0,
            "source": "security-advisories@github.com",
            "type": "Secondary"
          }
        ],
        "ssvcV203": [
          {
            "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
            "ssvcData": {
              "id": "CVE-2026-34938",
              "options": [
                {
                  "exploitation": "poc"
                },
                {
                  "automatable": "yes"
                },
                {
                  "technicalImpact": "total"
                }
              ],
              "role": "CISA Coordinator",
              "timestamp": "2026-04-06T13:23:32.754263Z",
              "version": "2.0.3"
            }
          }
        ]
      },
      "published": "2026-04-03T23:17:06.170",
      "references": [
        {
          "source": "security-advisories@github.com",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-6vh2-h83c-9294"
        }
      ],
      "sourceIdentifier": "security-advisories@github.com",
      "vulnStatus": "Analyzed",
      "weaknesses": [
        {
          "description": [
            {
              "lang": "en",
              "value": "CWE-693"
            }
          ],
          "source": "security-advisories@github.com",
          "type": "Secondary"
        }
      ]
    }

    FKIE_CVE-2026-34937

    Vulnerability from fkie_nvd - Published: 2026-04-03 23:17 - Updated: 2026-07-24 22:10
    Summary
    PraisonAI is a multi-agent teams system. Prior to version 1.5.90, run_python() in praisonai constructs a shell command string by interpolating user-controlled code into python3 -c "<code>" and passing it to subprocess.run(..., shell=True). The escaping logic only handles \ and ", leaving $() and backtick substitutions unescaped, allowing arbitrary OS command execution before Python is invoked. This issue has been patched in version 1.5.90.
    Impacted products
    Vendor Product Version
    praison praisonaiagents *

    {
      "affected": [
        {
          "affectedData": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.90"
                }
              ]
            }
          ],
          "source": "security-advisories@github.com"
        }
      ],
      "configurations": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:a:praison:praisonaiagents:*:*:*:*:*:*:*:*",
                  "matchCriteriaId": "0CECD474-C348-4AD9-B570-42895741C580",
                  "versionEndExcluding": "1.5.90",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "cveTags": [],
      "descriptions": [
        {
          "lang": "en",
          "value": "PraisonAI is a multi-agent teams system. Prior to version 1.5.90, run_python() in praisonai constructs a shell command string by interpolating user-controlled code into python3 -c \"\u003ccode\u003e\" and passing it to subprocess.run(..., shell=True). The escaping logic only handles \\ and \", leaving $() and backtick substitutions unescaped, allowing arbitrary OS command execution before Python is invoked. This issue has been patched in version 1.5.90."
        },
        {
          "lang": "es",
          "value": "PraisonAI es un sistema de equipos multiagente. Antes de la versi\u00f3n 1.5.90, run_python() en praisonai construye una cadena de comando de shell interpolando c\u00f3digo controlado por el usuario en python3 -c \u0027\u003ccode\u003e\u0027 y pas\u00e1ndolo a subprocess.run(..., shell=True). La l\u00f3gica de escape solo maneja \\ y \u0027, dejando las sustituciones de $() y de comillas invertidas sin escapar, lo que permite la ejecuci\u00f3n arbitraria de comandos del sistema operativo antes de que se invoque Python. Este problema ha sido parcheado en la versi\u00f3n 1.5.90."
        }
      ],
      "id": "CVE-2026-34937",
      "lastModified": "2026-07-24T22:10:00.140",
      "metrics": {
        "cvssMetricV31": [
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "LOCAL",
              "availabilityImpact": "HIGH",
              "baseScore": 7.8,
              "baseSeverity": "HIGH",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "HIGH",
              "privilegesRequired": "LOW",
              "scope": "UNCHANGED",
              "userInteraction": "NONE",
              "vectorString": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H",
              "version": "3.1"
            },
            "exploitabilityScore": 1.8,
            "impactScore": 5.9,
            "source": "security-advisories@github.com",
            "type": "Secondary"
          },
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "NETWORK",
              "availabilityImpact": "HIGH",
              "baseScore": 9.8,
              "baseSeverity": "CRITICAL",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "HIGH",
              "privilegesRequired": "NONE",
              "scope": "UNCHANGED",
              "userInteraction": "NONE",
              "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
              "version": "3.1"
            },
            "exploitabilityScore": 3.9,
            "impactScore": 5.9,
            "source": "nvd@nist.gov",
            "type": "Primary"
          }
        ],
        "ssvcV203": [
          {
            "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
            "ssvcData": {
              "id": "CVE-2026-34937",
              "options": [
                {
                  "exploitation": "poc"
                },
                {
                  "automatable": "no"
                },
                {
                  "technicalImpact": "total"
                }
              ],
              "role": "CISA Coordinator",
              "timestamp": "2026-04-06T16:08:04.394983Z",
              "version": "2.0.3"
            }
          }
        ]
      },
      "published": "2026-04-03T23:17:06.020",
      "references": [
        {
          "source": "security-advisories@github.com",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-w37c-qqfp-c67f"
        },
        {
          "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-w37c-qqfp-c67f"
        }
      ],
      "sourceIdentifier": "security-advisories@github.com",
      "vulnStatus": "Analyzed",
      "weaknesses": [
        {
          "description": [
            {
              "lang": "en",
              "value": "CWE-78"
            }
          ],
          "source": "security-advisories@github.com",
          "type": "Secondary"
        }
      ]
    }

    FKIE_CVE-2026-34954

    Vulnerability from fkie_nvd - Published: 2026-04-03 23:17 - Updated: 2026-07-24 21:10
    Summary
    PraisonAI is a multi-agent teams system. Prior to version 1.5.95, FileTools.download_file() in praisonaiagents validates the destination path but performs no validation on the url parameter, passing it directly to httpx.stream() with follow_redirects=True. An attacker who controls the URL can reach any host accessible from the server including cloud metadata services and internal network services. This issue has been patched in version 1.5.95.
    Impacted products
    Vendor Product Version
    praison praisonaiagents *

    {
      "affected": [
        {
          "affectedData": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.95"
                }
              ]
            }
          ],
          "source": "security-advisories@github.com"
        }
      ],
      "configurations": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:a:praison:praisonaiagents:*:*:*:*:*:*:*:*",
                  "matchCriteriaId": "549FBB43-121A-41E9-B6E6-FC14EC23A7B3",
                  "versionEndExcluding": "1.5.95",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "cveTags": [],
      "descriptions": [
        {
          "lang": "en",
          "value": "PraisonAI is a multi-agent teams system. Prior to version 1.5.95, FileTools.download_file() in praisonaiagents validates the destination path but performs no validation on the url parameter, passing it directly to httpx.stream() with follow_redirects=True. An attacker who controls the URL can reach any host accessible from the server including cloud metadata services and internal network services. This issue has been patched in version 1.5.95."
        },
        {
          "lang": "es",
          "value": "PraisonAI es un sistema de equipos multiagente. Antes de la versi\u00f3n 1.5.95, FileTools.download_file() en praisonaiagents valida la ruta de destino pero no realiza ninguna validaci\u00f3n en el par\u00e1metro url, pas\u00e1ndolo directamente a httpx.stream() con follow_redirects=True. Un atacante que controla la URL puede alcanzar cualquier host accesible desde el servidor, incluyendo servicios de metadatos en la nube y servicios de red internos. Este problema ha sido parcheado en la versi\u00f3n 1.5.95."
        }
      ],
      "id": "CVE-2026-34954",
      "lastModified": "2026-07-24T21:10:00.143",
      "metrics": {
        "cvssMetricV31": [
          {
            "cvssData": {
              "attackComplexity": "LOW",
              "attackVector": "NETWORK",
              "availabilityImpact": "NONE",
              "baseScore": 8.6,
              "baseSeverity": "HIGH",
              "confidentialityImpact": "HIGH",
              "integrityImpact": "NONE",
              "privilegesRequired": "NONE",
              "scope": "CHANGED",
              "userInteraction": "NONE",
              "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:N/A:N",
              "version": "3.1"
            },
            "exploitabilityScore": 3.9,
            "impactScore": 4.0,
            "source": "security-advisories@github.com",
            "type": "Secondary"
          }
        ],
        "ssvcV203": [
          {
            "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
            "ssvcData": {
              "id": "CVE-2026-34954",
              "options": [
                {
                  "exploitation": "poc"
                },
                {
                  "automatable": "yes"
                },
                {
                  "technicalImpact": "partial"
                }
              ],
              "role": "CISA Coordinator",
              "timestamp": "2026-04-06T13:22:54.634599Z",
              "version": "2.0.3"
            }
          }
        ]
      },
      "published": "2026-04-03T23:17:06.810",
      "references": [
        {
          "source": "security-advisories@github.com",
          "tags": [
            "Exploit",
            "Vendor Advisory"
          ],
          "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-44c2-3rw4-5gvh"
        }
      ],
      "sourceIdentifier": "security-advisories@github.com",
      "vulnStatus": "Analyzed",
      "weaknesses": [
        {
          "description": [
            {
              "lang": "en",
              "value": "CWE-918"
            }
          ],
          "source": "security-advisories@github.com",
          "type": "Secondary"
        }
      ]
    }

    CVE-2026-44339 (GCVE-0-2026-44339)

    Vulnerability from nvd – Published: 2026-05-08 13:37 – Updated: 2026-05-08 21:27
    VLAI
    Title
    PraisonAI has unsafe tool resolution in `ToolExecutionMixin.execute_tool`: undeclared `__main__` callables execute
    Summary
    PraisonAI is a multi-agent teams system. Prior to praisonai version 4.6.37 and praisonaiagents version 1.6.37, praisonaiagents resolves unresolved tool names against module globals and __main__ after it fails to match the declared tool list and the registry. With the default agent configuration, _perm_allow is None, so undeclared non-dangerous tool names are not rejected by the permission gate. An attacker who can influence tool-call names can therefore invoke unintended application callables that were never declared as tools. This issue has been patched in praisonai version 4.6.37 and praisonaiagents version 1.6.37.
    SSVC
    Exploitation: poc Automatable: yes Technical Impact: partial
    CISA Coordinator · CISA-ADP (v2.0.3)
    Decision recorded 2026-05-08 17:03 UTC
    CWE
    • CWE-470 - Use of Externally-Controlled Input to Select Classes or Code ('Unsafe Reflection')
    Assigner
    GitHub_M CNA under the mitre root
    CNA scorecard C 72/100 over 8156 records in the last 180 days details
    References
    Impacted products
    Vendor Product Version
    MervinPraison PraisonAI Affected: praisonaiagents < 1.6.37
    Affected: praisonai < 4.6.37
    Create a notification for this product.
    Show details on NVD website

    {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-44339",
                    "options": [
                      {
                        "Exploitation": "poc"
                      },
                      {
                        "Automatable": "yes"
                      },
                      {
                        "Technical Impact": "partial"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-05-08T17:03:56.082162Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-05-08T21:27:22.924Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "references": [
              {
                "tags": [
                  "exploit"
                ],
                "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-gmjg-hv98-qggq"
              }
            ],
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "praisonaiagents \u003c 1.6.37"
                },
                {
                  "status": "affected",
                  "version": "praisonai \u003c 4.6.37"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "PraisonAI is a multi-agent teams system. Prior to praisonai version 4.6.37 and praisonaiagents version 1.6.37, praisonaiagents resolves unresolved tool names against module globals and __main__ after it fails to match the declared tool list and the registry. With the default agent configuration, _perm_allow is None, so undeclared non-dangerous tool names are not rejected by the permission gate. An attacker who can influence tool-call names can therefore invoke unintended application callables that were never declared as tools. This issue has been patched in praisonai version 4.6.37 and praisonaiagents version 1.6.37."
            }
          ],
          "metrics": [
            {
              "cvssV3_1": {
                "attackComplexity": "LOW",
                "attackVector": "NETWORK",
                "availabilityImpact": "LOW",
                "baseScore": 8.6,
                "baseSeverity": "HIGH",
                "confidentialityImpact": "LOW",
                "integrityImpact": "HIGH",
                "privilegesRequired": "NONE",
                "scope": "UNCHANGED",
                "userInteraction": "NONE",
                "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:H/A:L",
                "version": "3.1"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-470",
                  "description": "CWE-470: Use of Externally-Controlled Input to Select Classes or Code (\u0027Unsafe Reflection\u0027)",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-05-08T13:37:09.706Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-gmjg-hv98-qggq",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-gmjg-hv98-qggq"
            }
          ],
          "source": {
            "advisory": "GHSA-gmjg-hv98-qggq",
            "discovery": "UNKNOWN"
          },
          "title": "PraisonAI has unsafe tool resolution in `ToolExecutionMixin.execute_tool`: undeclared `__main__` callables execute"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-44339",
        "datePublished": "2026-05-08T13:37:09.706Z",
        "dateReserved": "2026-05-05T19:52:59.147Z",
        "dateUpdated": "2026-05-08T21:27:22.924Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }

    CVE-2026-44335 (GCVE-0-2026-44335)

    Vulnerability from nvd – Published: 2026-05-08 13:26 – Updated: 2026-05-08 14:46
    VLAI
    Title
    SSRF bypass in PraisonAI
    Summary
    PraisonAI is a multi-agent teams system. Prior to version 1.6.32, the URL checking logic in PraisonAI has a logical flaw that could be bypassed by attackers, leading to SSRF attacks. This issue has been patched in version 1.6.32.
    SSVC
    Exploitation: poc Automatable: yes Technical Impact: partial
    CISA Coordinator · CISA-ADP (v2.0.3)
    Decision recorded 2026-05-08 14:46 UTC
    CWE
    • CWE-918 - Server-Side Request Forgery (SSRF)
    Assigner
    GitHub_M CNA under the mitre root
    CNA scorecard C 72/100 over 8156 records in the last 180 days details
    References
    Impacted products
    Vendor Product Version
    MervinPraison PraisonAI Affected: < 1.6.32
    Create a notification for this product.
    Show details on NVD website

    {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-44335",
                    "options": [
                      {
                        "Exploitation": "poc"
                      },
                      {
                        "Automatable": "yes"
                      },
                      {
                        "Technical Impact": "partial"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-05-08T14:46:06.544307Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-05-08T14:46:13.605Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "references": [
              {
                "tags": [
                  "exploit"
                ],
                "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-q9pw-vmhh-384g"
              }
            ],
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.6.32"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "PraisonAI is a multi-agent teams system. Prior to version 1.6.32, the URL checking logic in PraisonAI has a logical flaw that could be bypassed by attackers, leading to SSRF attacks. This issue has been patched in version 1.6.32."
            }
          ],
          "metrics": [
            {
              "cvssV4_0": {
                "attackComplexity": "LOW",
                "attackRequirements": "NONE",
                "attackVector": "NETWORK",
                "baseScore": 7.7,
                "baseSeverity": "HIGH",
                "privilegesRequired": "NONE",
                "subAvailabilityImpact": "NONE",
                "subConfidentialityImpact": "NONE",
                "subIntegrityImpact": "NONE",
                "userInteraction": "NONE",
                "vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:N/SC:N/SI:N/SA:N/E:P",
                "version": "4.0",
                "vulnAvailabilityImpact": "NONE",
                "vulnConfidentialityImpact": "NONE",
                "vulnIntegrityImpact": "HIGH"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-918",
                  "description": "CWE-918: Server-Side Request Forgery (SSRF)",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-05-08T13:26:48.564Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-q9pw-vmhh-384g",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-q9pw-vmhh-384g"
            }
          ],
          "source": {
            "advisory": "GHSA-q9pw-vmhh-384g",
            "discovery": "UNKNOWN"
          },
          "title": "SSRF bypass in PraisonAI"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-44335",
        "datePublished": "2026-05-08T13:26:48.564Z",
        "dateReserved": "2026-05-05T19:52:59.147Z",
        "dateUpdated": "2026-05-08T14:46:13.605Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }

    CVE-2026-41496 (GCVE-0-2026-41496)

    Vulnerability from nvd – Published: 2026-05-08 13:19 – Updated: 2026-05-08 23:18
    VLAI
    Title
    PraisonAI: SQL Injection via unvalidated `table_prefix` in 9 conversation store backends (incomplete fix for CVE-2026-40315)
    Summary
    PraisonAI is a multi-agent teams system. Prior to praisonai version 4.6.9 and praisonaiagents version 1.6.9, the fix for CVE-2026-40315 added input validation to SQLiteConversationStore only. Nine sibling backends — MySQL, PostgreSQL, async SQLite/MySQL/PostgreSQL, Turso, SingleStore, Supabase, SurrealDB — pass table_prefix straight into f-string SQL. Same root cause, same code pattern, same exploitation. 52 unvalidated injection points across the codebase. postgres.py additionally accepts an unvalidated schema parameter used directly in DDL. This issue has been patched in praisonai version 4.6.9 and praisonaiagents version 1.6.9.
    SSVC
    Exploitation: poc Automatable: no Technical Impact: total
    CISA Coordinator · CISA-ADP (v2.0.3)
    Decision recorded 2026-05-08 23:17 UTC
    CWE
    • CWE-89 - Improper Neutralization of Special Elements used in an SQL Command ('SQL Injection')
    Assigner
    GitHub_M CNA under the mitre root
    CNA scorecard C 72/100 over 8156 records in the last 180 days details
    References
    Impacted products
    Vendor Product Version
    MervinPraison PraisonAI Affected: praisonaiagents < 1.6.9
    Affected: praisonai < 4.6.9
    Create a notification for this product.
    Show details on NVD website

    {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-41496",
                    "options": [
                      {
                        "Exploitation": "poc"
                      },
                      {
                        "Automatable": "no"
                      },
                      {
                        "Technical Impact": "total"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-05-08T23:17:23.691346Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-05-08T23:18:11.328Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "references": [
              {
                "tags": [
                  "exploit"
                ],
                "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-rg3h-x3jw-7jm5"
              }
            ],
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "praisonaiagents \u003c 1.6.9"
                },
                {
                  "status": "affected",
                  "version": "praisonai \u003c 4.6.9"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "PraisonAI is a multi-agent teams system. Prior to praisonai version 4.6.9 and praisonaiagents version 1.6.9, the fix for CVE-2026-40315 added input validation to SQLiteConversationStore only. Nine sibling backends \u2014 MySQL, PostgreSQL, async SQLite/MySQL/PostgreSQL, Turso, SingleStore, Supabase, SurrealDB \u2014 pass table_prefix straight into f-string SQL. Same root cause, same code pattern, same exploitation. 52 unvalidated injection points across the codebase. postgres.py additionally accepts an unvalidated schema parameter used directly in DDL. This issue has been patched in praisonai version 4.6.9 and praisonaiagents version 1.6.9."
            }
          ],
          "metrics": [
            {
              "cvssV3_1": {
                "attackComplexity": "LOW",
                "attackVector": "NETWORK",
                "availabilityImpact": "NONE",
                "baseScore": 8.1,
                "baseSeverity": "HIGH",
                "confidentialityImpact": "HIGH",
                "integrityImpact": "HIGH",
                "privilegesRequired": "LOW",
                "scope": "UNCHANGED",
                "userInteraction": "NONE",
                "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:N",
                "version": "3.1"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-89",
                  "description": "CWE-89: Improper Neutralization of Special Elements used in an SQL Command (\u0027SQL Injection\u0027)",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-05-08T13:19:10.753Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-rg3h-x3jw-7jm5",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-rg3h-x3jw-7jm5"
            }
          ],
          "source": {
            "advisory": "GHSA-rg3h-x3jw-7jm5",
            "discovery": "UNKNOWN"
          },
          "title": "PraisonAI: SQL Injection via unvalidated `table_prefix` in 9 conversation store backends (incomplete fix for CVE-2026-40315)"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-41496",
        "datePublished": "2026-05-08T13:19:10.753Z",
        "dateReserved": "2026-04-20T16:14:19.009Z",
        "dateUpdated": "2026-05-08T23:18:11.328Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }

    CVE-2026-40289 (GCVE-0-2026-40289)

    Vulnerability from nvd – Published: 2026-04-14 03:05 – Updated: 2026-04-14 20:18
    VLAI
    Title
    PraisonAI Browser Server allows unauthenticated WebSocket clients to hijack connected extension sessions
    Summary
    PraisonAI is a multi-agent teams system. In versions below 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents, the browser bridge (praisonai browser start) is vulnerable to unauthenticated remote session hijacking due to missing authentication and a bypassable origin check on its /ws WebSocket endpoint. The server binds to 0.0.0.0 by default and only validates the Origin header when one is present, meaning any non-browser client that omits the header is accepted without restriction. An unauthenticated network attacker can connect, send a start_session message, and the server will route it to the first idle browser-extension WebSocket (effectively hijacking that session) and then broadcast all resulting automation actions and outputs back to the attacker. This enables unauthorized remote control of connected browser automation sessions, leakage of sensitive page context and automation results, and misuse of model-backed browser actions in any environment where the bridge is network-reachable. This issue has been fixed in versions 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents.
    SSVC
    Exploitation: poc Automatable: yes Technical Impact: total
    CISA Coordinator · CISA-ADP (v2.0.3)
    Decision recorded 2026-04-14 20:18 UTC
    CWE
    • CWE-306 - Missing Authentication for Critical Function
    Assigner
    GitHub_M CNA under the mitre root
    CNA scorecard C 72/100 over 8156 records in the last 180 days details
    References
    Impacted products
    Show details on NVD website

    {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-40289",
                    "options": [
                      {
                        "Exploitation": "poc"
                      },
                      {
                        "Automatable": "yes"
                      },
                      {
                        "Technical Impact": "total"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-04-14T20:18:27.623888Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-04-14T20:18:37.319Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 4.5.139"
                }
              ]
            },
            {
              "product": "praisonaiagents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.140"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "PraisonAI is a multi-agent teams system. In versions below 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents, the browser bridge (praisonai browser start) is vulnerable to unauthenticated remote session hijacking due to missing authentication and a bypassable origin check on its /ws WebSocket endpoint. The server binds to 0.0.0.0 by default and only validates the Origin header when one is present, meaning any non-browser client that omits the header is accepted without restriction. An unauthenticated network attacker can connect, send a start_session message, and the server will route it to the first idle browser-extension WebSocket (effectively hijacking that session) and then broadcast all resulting automation actions and outputs back to the attacker. This enables unauthorized remote control of connected browser automation sessions, leakage of sensitive page context and automation results, and misuse of model-backed browser actions in any environment where the bridge is network-reachable. This issue has been fixed in versions 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents."
            }
          ],
          "metrics": [
            {
              "cvssV3_1": {
                "attackComplexity": "LOW",
                "attackVector": "NETWORK",
                "availabilityImpact": "NONE",
                "baseScore": 9.1,
                "baseSeverity": "CRITICAL",
                "confidentialityImpact": "HIGH",
                "integrityImpact": "HIGH",
                "privilegesRequired": "NONE",
                "scope": "UNCHANGED",
                "userInteraction": "NONE",
                "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N",
                "version": "3.1"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-306",
                  "description": "CWE-306: Missing Authentication for Critical Function",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-04-14T03:05:05.514Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-8x8f-54wf-vv92",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-8x8f-54wf-vv92"
            }
          ],
          "source": {
            "advisory": "GHSA-8x8f-54wf-vv92",
            "discovery": "UNKNOWN"
          },
          "title": "PraisonAI Browser Server allows unauthenticated WebSocket clients to hijack connected extension sessions"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-40289",
        "datePublished": "2026-04-14T03:05:05.514Z",
        "dateReserved": "2026-04-10T20:22:44.035Z",
        "dateUpdated": "2026-04-14T20:18:37.319Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }

    CVE-2026-40288 (GCVE-0-2026-40288)

    Vulnerability from nvd – Published: 2026-04-14 03:00 – Updated: 2026-04-14 15:57
    VLAI
    Title
    PraisonAI: Critical RCE via `type: job` workflow YAML
    Summary
    PraisonAI is a multi-agent teams system. In versions below 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents, the workflow engine is vulnerable to arbitrary command and code execution through untrusted YAML files. When praisonai workflow run <file.yaml> loads a YAML file with type: job, the JobWorkflowExecutor in job_workflow.py processes steps that support run: (shell commands via subprocess.run()), script: (inline Python via exec()), and python: (arbitrary Python script execution)—all without any validation, sandboxing, or user confirmation. The affected code paths include action_run() in workflow.py and _exec_shell(), _exec_inline_python(), and _exec_python_script() in job_workflow.py. An attacker who can supply or influence a workflow YAML file (particularly in CI pipelines, shared repositories, or multi-tenant deployment environments) can achieve full arbitrary command execution on the host system, compromising the machine and any accessible data or credentials. This issue has been fixed in versions 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents.
    SSVC
    Exploitation: poc Automatable: yes Technical Impact: total
    CISA Coordinator · CISA-ADP (v2.0.3)
    Decision recorded 2026-04-14 15:56 UTC
    CWE
    • CWE-78 - Improper Neutralization of Special Elements used in an OS Command ('OS Command Injection')
    • CWE-94 - Improper Control of Generation of Code ('Code Injection')
    Assigner
    GitHub_M CNA under the mitre root
    CNA scorecard C 72/100 over 8156 records in the last 180 days details
    References
    Impacted products
    Show details on NVD website

    {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-40288",
                    "options": [
                      {
                        "Exploitation": "poc"
                      },
                      {
                        "Automatable": "yes"
                      },
                      {
                        "Technical Impact": "total"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-04-14T15:56:49.344666Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-04-14T15:57:15.812Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 4.5.139"
                }
              ]
            },
            {
              "product": "praisonaiagents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.140"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "PraisonAI is a multi-agent teams system. In versions below 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents, the workflow engine is vulnerable to arbitrary command and code execution through untrusted YAML files. When praisonai workflow run \u003cfile.yaml\u003e loads a YAML file with type: job, the JobWorkflowExecutor in job_workflow.py processes steps that support run: (shell commands via subprocess.run()), script: (inline Python via exec()), and python: (arbitrary Python script execution)\u2014all without any validation, sandboxing, or user confirmation. The affected code paths include action_run() in workflow.py and _exec_shell(), _exec_inline_python(), and _exec_python_script() in job_workflow.py. An attacker who can supply or influence a workflow YAML file (particularly in CI pipelines, shared repositories, or multi-tenant deployment environments) can achieve full arbitrary command execution on the host system, compromising the machine and any accessible data or credentials. This issue has been fixed in versions 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents."
            }
          ],
          "metrics": [
            {
              "cvssV3_1": {
                "attackComplexity": "LOW",
                "attackVector": "NETWORK",
                "availabilityImpact": "HIGH",
                "baseScore": 9.8,
                "baseSeverity": "CRITICAL",
                "confidentialityImpact": "HIGH",
                "integrityImpact": "HIGH",
                "privilegesRequired": "NONE",
                "scope": "UNCHANGED",
                "userInteraction": "NONE",
                "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
                "version": "3.1"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-78",
                  "description": "CWE-78: Improper Neutralization of Special Elements used in an OS Command (\u0027OS Command Injection\u0027)",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            },
            {
              "descriptions": [
                {
                  "cweId": "CWE-94",
                  "description": "CWE-94: Improper Control of Generation of Code (\u0027Code Injection\u0027)",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-04-14T03:01:53.412Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-vc46-vw85-3wvm",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-vc46-vw85-3wvm"
            }
          ],
          "source": {
            "advisory": "GHSA-vc46-vw85-3wvm",
            "discovery": "UNKNOWN"
          },
          "title": "PraisonAI: Critical RCE via `type: job` workflow YAML"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-40288",
        "datePublished": "2026-04-14T03:00:21.547Z",
        "dateReserved": "2026-04-10T20:22:44.035Z",
        "dateUpdated": "2026-04-14T15:57:15.812Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }

    CVE-2026-40287 (GCVE-0-2026-40287)

    Vulnerability from nvd – Published: 2026-04-14 02:55 – Updated: 2026-04-14 13:23
    VLAI
    Title
    PraisonAI has RCE via Automatic tools.py Import
    Summary
    PraisonAI is a multi-agent teams system. Versions 4.5.138 and below are vulnerable to arbitrary code execution through automatic, unsanitized import of a tools.py file from the current working directory. Components including call.py (import_tools_from_file()), tool_resolver.py (_load_local_tools()), and CLI tool-loading paths blindly import ./tools.py at startup without any validation, sandboxing, or user confirmation. An attacker who can place a malicious tools.py in the directory where PraisonAI is launched (such as through a shared project, cloned repository, or writable workspace) achieves immediate arbitrary Python code execution in the host environment. This compromises the full PraisonAI process, the host system, and any connected data or credentials. This issue has been fixed in version 4.5.139.
    SSVC
    Exploitation: poc Automatable: no Technical Impact: total
    CISA Coordinator · CISA-ADP (v2.0.3)
    Decision recorded 2026-04-14 13:23 UTC
    CWE
    • CWE-94 - Improper Control of Generation of Code ('Code Injection')
    • CWE-426 - Untrusted Search Path
    Assigner
    GitHub_M CNA under the mitre root
    CNA scorecard C 72/100 over 8156 records in the last 180 days details
    References
    Impacted products
    Show details on NVD website

    {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-40287",
                    "options": [
                      {
                        "Exploitation": "poc"
                      },
                      {
                        "Automatable": "no"
                      },
                      {
                        "Technical Impact": "total"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-04-14T13:23:23.221160Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-04-14T13:23:29.807Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "references": [
              {
                "tags": [
                  "exploit"
                ],
                "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-g985-wjh9-qxxc"
              }
            ],
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 4.5.139"
                }
              ]
            },
            {
              "product": "praisonaiagents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.140"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "PraisonAI is a multi-agent teams system. Versions 4.5.138 and below are vulnerable to arbitrary code execution through automatic, unsanitized import of a tools.py file from the current working directory. Components including call.py (import_tools_from_file()), tool_resolver.py (_load_local_tools()), and CLI tool-loading paths blindly import ./tools.py at startup without any validation, sandboxing, or user confirmation. An attacker who can place a malicious tools.py in the directory where PraisonAI is launched (such as through a shared project, cloned repository, or writable workspace) achieves immediate arbitrary Python code execution in the host environment. This compromises the full PraisonAI process, the host system, and any connected data or credentials. This issue has been fixed in version 4.5.139."
            }
          ],
          "metrics": [
            {
              "cvssV3_1": {
                "attackComplexity": "LOW",
                "attackVector": "LOCAL",
                "availabilityImpact": "HIGH",
                "baseScore": 8.4,
                "baseSeverity": "HIGH",
                "confidentialityImpact": "HIGH",
                "integrityImpact": "HIGH",
                "privilegesRequired": "NONE",
                "scope": "UNCHANGED",
                "userInteraction": "NONE",
                "vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
                "version": "3.1"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-94",
                  "description": "CWE-94: Improper Control of Generation of Code (\u0027Code Injection\u0027)",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            },
            {
              "descriptions": [
                {
                  "cweId": "CWE-426",
                  "description": "CWE-426: Untrusted Search Path",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-04-14T02:55:38.270Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-g985-wjh9-qxxc",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-g985-wjh9-qxxc"
            }
          ],
          "source": {
            "advisory": "GHSA-g985-wjh9-qxxc",
            "discovery": "UNKNOWN"
          },
          "title": "PraisonAI has RCE via Automatic tools.py Import"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-40287",
        "datePublished": "2026-04-14T02:55:38.270Z",
        "dateReserved": "2026-04-10T20:22:44.035Z",
        "dateUpdated": "2026-04-14T13:23:29.807Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }

    CVE-2026-40160 (GCVE-0-2026-40160)

    Vulnerability from nvd – Published: 2026-04-10 16:59 – Updated: 2026-04-13 15:36
    VLAI
    Title
    PraisonAIAgents has SSRF via unvalidated URL in `web_crawl` httpx fallback
    Summary
    PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, web_crawl's httpx fallback path passes user-supplied URLs directly to httpx.AsyncClient.get() with follow_redirects=True and no host validation. An LLM agent tricked into crawling an internal URL can reach cloud metadata endpoints (169.254.169.254), internal services, and localhost. The response content is returned to the agent and may appear in output visible to the attacker. This fallback is the default crawl path on a fresh PraisonAI installation (no Tavily key, no Crawl4AI installed). This vulnerability is fixed in 1.5.128.
    SSVC
    Exploitation: poc Automatable: no Technical Impact: partial
    CISA Coordinator · CISA-ADP (v2.0.3)
    Decision recorded 2026-04-13 15:28 UTC
    CWE
    • CWE-918 - Server-Side Request Forgery (SSRF)
    Assigner
    GitHub_M CNA under the mitre root
    CNA scorecard C 72/100 over 8156 records in the last 180 days details
    References
    Impacted products
    Show details on NVD website

    {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-40160",
                    "options": [
                      {
                        "Exploitation": "poc"
                      },
                      {
                        "Automatable": "no"
                      },
                      {
                        "Technical Impact": "partial"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-04-13T15:28:31.603808Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-04-13T15:36:41.165Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "references": [
              {
                "tags": [
                  "exploit"
                ],
                "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-qq9r-63f6-v542"
              }
            ],
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "PraisonAIAgents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.128"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, web_crawl\u0027s httpx fallback path passes user-supplied URLs directly to httpx.AsyncClient.get() with follow_redirects=True and no host validation. An LLM agent tricked into crawling an internal URL can reach cloud metadata endpoints (169.254.169.254), internal services, and localhost. The response content is returned to the agent and may appear in output visible to the attacker. This fallback is the default crawl path on a fresh PraisonAI installation (no Tavily key, no Crawl4AI installed). This vulnerability is fixed in 1.5.128."
            }
          ],
          "metrics": [
            {
              "cvssV4_0": {
                "attackComplexity": "LOW",
                "attackRequirements": "PRESENT",
                "attackVector": "NETWORK",
                "baseScore": 7.1,
                "baseSeverity": "HIGH",
                "privilegesRequired": "NONE",
                "subAvailabilityImpact": "NONE",
                "subConfidentialityImpact": "HIGH",
                "subIntegrityImpact": "LOW",
                "userInteraction": "PASSIVE",
                "vectorString": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:P/VC:H/VI:N/VA:N/SC:H/SI:L/SA:N",
                "version": "4.0",
                "vulnAvailabilityImpact": "NONE",
                "vulnConfidentialityImpact": "HIGH",
                "vulnIntegrityImpact": "NONE"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-918",
                  "description": "CWE-918: Server-Side Request Forgery (SSRF)",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-04-10T16:59:09.542Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-qq9r-63f6-v542",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-qq9r-63f6-v542"
            }
          ],
          "source": {
            "advisory": "GHSA-qq9r-63f6-v542",
            "discovery": "UNKNOWN"
          },
          "title": "PraisonAIAgents has SSRF via unvalidated URL in `web_crawl` httpx fallback"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-40160",
        "datePublished": "2026-04-10T16:59:09.542Z",
        "dateReserved": "2026-04-09T19:31:56.014Z",
        "dateUpdated": "2026-04-13T15:36:41.165Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }

    CVE-2026-40153 (GCVE-0-2026-40153)

    Vulnerability from nvd – Published: 2026-04-09 21:27 – Updated: 2026-04-13 15:37
    VLAI
    Title
    PraisonAIAgents Affected by Environment Variable Secret Exfiltration via os.path.expandvars() Bypassing shell=False in Shell Tool
    Summary
    PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, the execute_command function in shell_tools.py calls os.path.expandvars() on every command argument at line 64, manually re-implementing shell-level environment variable expansion despite using shell=False (line 88) for security. This allows exfiltration of secrets stored in environment variables (database credentials, API keys, cloud access keys). The approval system displays the unexpanded $VAR references to human reviewers, creating a deceptive approval where the displayed command differs from what actually executes. This vulnerability is fixed in 1.5.128.
    SSVC
    Exploitation: poc Automatable: no Technical Impact: partial
    CISA Coordinator · CISA-ADP (v2.0.3)
    Decision recorded 2026-04-13 15:28 UTC
    CWE
    • CWE-526 - Cleartext Storage of Sensitive Information in an Environment Variable
    Assigner
    GitHub_M CNA under the mitre root
    CNA scorecard C 72/100 over 8156 records in the last 180 days details
    References
    Impacted products
    Show details on NVD website

    {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-40153",
                    "options": [
                      {
                        "Exploitation": "poc"
                      },
                      {
                        "Automatable": "no"
                      },
                      {
                        "Technical Impact": "partial"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-04-13T15:28:34.665913Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-04-13T15:37:40.235Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "references": [
              {
                "tags": [
                  "exploit"
                ],
                "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-v8g7-9q6v-p3x8"
              }
            ],
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "PraisonAIAgents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.128"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, the execute_command function in shell_tools.py calls os.path.expandvars() on every command argument at line 64, manually re-implementing shell-level environment variable expansion despite using shell=False (line 88) for security. This allows exfiltration of secrets stored in environment variables (database credentials, API keys, cloud access keys). The approval system displays the unexpanded $VAR references to human reviewers, creating a deceptive approval where the displayed command differs from what actually executes. This vulnerability is fixed in 1.5.128."
            }
          ],
          "metrics": [
            {
              "cvssV3_1": {
                "attackComplexity": "LOW",
                "attackVector": "NETWORK",
                "availabilityImpact": "NONE",
                "baseScore": 7.4,
                "baseSeverity": "HIGH",
                "confidentialityImpact": "HIGH",
                "integrityImpact": "NONE",
                "privilegesRequired": "NONE",
                "scope": "CHANGED",
                "userInteraction": "REQUIRED",
                "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:N/A:N",
                "version": "3.1"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-526",
                  "description": "CWE-526: Cleartext Storage of Sensitive Information in an Environment Variable",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-04-09T21:27:45.545Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-v8g7-9q6v-p3x8",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-v8g7-9q6v-p3x8"
            }
          ],
          "source": {
            "advisory": "GHSA-v8g7-9q6v-p3x8",
            "discovery": "UNKNOWN"
          },
          "title": "PraisonAIAgents Affected by Environment Variable Secret Exfiltration via os.path.expandvars() Bypassing shell=False in Shell Tool"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-40153",
        "datePublished": "2026-04-09T21:27:45.545Z",
        "dateReserved": "2026-04-09T19:31:56.013Z",
        "dateUpdated": "2026-04-13T15:37:40.235Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }

    CVE-2026-44339 (GCVE-0-2026-44339)

    Vulnerability from cvelistv5 – Published: 2026-05-08 13:37 – Updated: 2026-05-08 21:27
    VLAI
    Title
    PraisonAI has unsafe tool resolution in `ToolExecutionMixin.execute_tool`: undeclared `__main__` callables execute
    Summary
    PraisonAI is a multi-agent teams system. Prior to praisonai version 4.6.37 and praisonaiagents version 1.6.37, praisonaiagents resolves unresolved tool names against module globals and __main__ after it fails to match the declared tool list and the registry. With the default agent configuration, _perm_allow is None, so undeclared non-dangerous tool names are not rejected by the permission gate. An attacker who can influence tool-call names can therefore invoke unintended application callables that were never declared as tools. This issue has been patched in praisonai version 4.6.37 and praisonaiagents version 1.6.37.
    SSVC
    Exploitation: poc Automatable: yes Technical Impact: partial
    CISA Coordinator · CISA-ADP (v2.0.3)
    Decision recorded 2026-05-08 17:03 UTC
    CWE
    • CWE-470 - Use of Externally-Controlled Input to Select Classes or Code ('Unsafe Reflection')
    Assigner
    GitHub_M CNA under the mitre root
    CNA scorecard C 72/100 over 8156 records in the last 180 days details
    References
    Impacted products
    Vendor Product Version
    MervinPraison PraisonAI Affected: praisonaiagents < 1.6.37
    Affected: praisonai < 4.6.37
    Create a notification for this product.
    Show details on NVD website

    {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-44339",
                    "options": [
                      {
                        "Exploitation": "poc"
                      },
                      {
                        "Automatable": "yes"
                      },
                      {
                        "Technical Impact": "partial"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-05-08T17:03:56.082162Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-05-08T21:27:22.924Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "references": [
              {
                "tags": [
                  "exploit"
                ],
                "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-gmjg-hv98-qggq"
              }
            ],
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "praisonaiagents \u003c 1.6.37"
                },
                {
                  "status": "affected",
                  "version": "praisonai \u003c 4.6.37"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "PraisonAI is a multi-agent teams system. Prior to praisonai version 4.6.37 and praisonaiagents version 1.6.37, praisonaiagents resolves unresolved tool names against module globals and __main__ after it fails to match the declared tool list and the registry. With the default agent configuration, _perm_allow is None, so undeclared non-dangerous tool names are not rejected by the permission gate. An attacker who can influence tool-call names can therefore invoke unintended application callables that were never declared as tools. This issue has been patched in praisonai version 4.6.37 and praisonaiagents version 1.6.37."
            }
          ],
          "metrics": [
            {
              "cvssV3_1": {
                "attackComplexity": "LOW",
                "attackVector": "NETWORK",
                "availabilityImpact": "LOW",
                "baseScore": 8.6,
                "baseSeverity": "HIGH",
                "confidentialityImpact": "LOW",
                "integrityImpact": "HIGH",
                "privilegesRequired": "NONE",
                "scope": "UNCHANGED",
                "userInteraction": "NONE",
                "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:H/A:L",
                "version": "3.1"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-470",
                  "description": "CWE-470: Use of Externally-Controlled Input to Select Classes or Code (\u0027Unsafe Reflection\u0027)",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-05-08T13:37:09.706Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-gmjg-hv98-qggq",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-gmjg-hv98-qggq"
            }
          ],
          "source": {
            "advisory": "GHSA-gmjg-hv98-qggq",
            "discovery": "UNKNOWN"
          },
          "title": "PraisonAI has unsafe tool resolution in `ToolExecutionMixin.execute_tool`: undeclared `__main__` callables execute"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-44339",
        "datePublished": "2026-05-08T13:37:09.706Z",
        "dateReserved": "2026-05-05T19:52:59.147Z",
        "dateUpdated": "2026-05-08T21:27:22.924Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }

    CVE-2026-44335 (GCVE-0-2026-44335)

    Vulnerability from cvelistv5 – Published: 2026-05-08 13:26 – Updated: 2026-05-08 14:46
    VLAI
    Title
    SSRF bypass in PraisonAI
    Summary
    PraisonAI is a multi-agent teams system. Prior to version 1.6.32, the URL checking logic in PraisonAI has a logical flaw that could be bypassed by attackers, leading to SSRF attacks. This issue has been patched in version 1.6.32.
    SSVC
    Exploitation: poc Automatable: yes Technical Impact: partial
    CISA Coordinator · CISA-ADP (v2.0.3)
    Decision recorded 2026-05-08 14:46 UTC
    CWE
    • CWE-918 - Server-Side Request Forgery (SSRF)
    Assigner
    GitHub_M CNA under the mitre root
    CNA scorecard C 72/100 over 8156 records in the last 180 days details
    References
    Impacted products
    Vendor Product Version
    MervinPraison PraisonAI Affected: < 1.6.32
    Create a notification for this product.
    Show details on NVD website

    {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-44335",
                    "options": [
                      {
                        "Exploitation": "poc"
                      },
                      {
                        "Automatable": "yes"
                      },
                      {
                        "Technical Impact": "partial"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-05-08T14:46:06.544307Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-05-08T14:46:13.605Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "references": [
              {
                "tags": [
                  "exploit"
                ],
                "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-q9pw-vmhh-384g"
              }
            ],
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.6.32"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "PraisonAI is a multi-agent teams system. Prior to version 1.6.32, the URL checking logic in PraisonAI has a logical flaw that could be bypassed by attackers, leading to SSRF attacks. This issue has been patched in version 1.6.32."
            }
          ],
          "metrics": [
            {
              "cvssV4_0": {
                "attackComplexity": "LOW",
                "attackRequirements": "NONE",
                "attackVector": "NETWORK",
                "baseScore": 7.7,
                "baseSeverity": "HIGH",
                "privilegesRequired": "NONE",
                "subAvailabilityImpact": "NONE",
                "subConfidentialityImpact": "NONE",
                "subIntegrityImpact": "NONE",
                "userInteraction": "NONE",
                "vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:N/SC:N/SI:N/SA:N/E:P",
                "version": "4.0",
                "vulnAvailabilityImpact": "NONE",
                "vulnConfidentialityImpact": "NONE",
                "vulnIntegrityImpact": "HIGH"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-918",
                  "description": "CWE-918: Server-Side Request Forgery (SSRF)",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-05-08T13:26:48.564Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-q9pw-vmhh-384g",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-q9pw-vmhh-384g"
            }
          ],
          "source": {
            "advisory": "GHSA-q9pw-vmhh-384g",
            "discovery": "UNKNOWN"
          },
          "title": "SSRF bypass in PraisonAI"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-44335",
        "datePublished": "2026-05-08T13:26:48.564Z",
        "dateReserved": "2026-05-05T19:52:59.147Z",
        "dateUpdated": "2026-05-08T14:46:13.605Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }

    CVE-2026-41496 (GCVE-0-2026-41496)

    Vulnerability from cvelistv5 – Published: 2026-05-08 13:19 – Updated: 2026-05-08 23:18
    VLAI
    Title
    PraisonAI: SQL Injection via unvalidated `table_prefix` in 9 conversation store backends (incomplete fix for CVE-2026-40315)
    Summary
    PraisonAI is a multi-agent teams system. Prior to praisonai version 4.6.9 and praisonaiagents version 1.6.9, the fix for CVE-2026-40315 added input validation to SQLiteConversationStore only. Nine sibling backends — MySQL, PostgreSQL, async SQLite/MySQL/PostgreSQL, Turso, SingleStore, Supabase, SurrealDB — pass table_prefix straight into f-string SQL. Same root cause, same code pattern, same exploitation. 52 unvalidated injection points across the codebase. postgres.py additionally accepts an unvalidated schema parameter used directly in DDL. This issue has been patched in praisonai version 4.6.9 and praisonaiagents version 1.6.9.
    SSVC
    Exploitation: poc Automatable: no Technical Impact: total
    CISA Coordinator · CISA-ADP (v2.0.3)
    Decision recorded 2026-05-08 23:17 UTC
    CWE
    • CWE-89 - Improper Neutralization of Special Elements used in an SQL Command ('SQL Injection')
    Assigner
    GitHub_M CNA under the mitre root
    CNA scorecard C 72/100 over 8156 records in the last 180 days details
    References
    Impacted products
    Vendor Product Version
    MervinPraison PraisonAI Affected: praisonaiagents < 1.6.9
    Affected: praisonai < 4.6.9
    Create a notification for this product.
    Show details on NVD website

    {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-41496",
                    "options": [
                      {
                        "Exploitation": "poc"
                      },
                      {
                        "Automatable": "no"
                      },
                      {
                        "Technical Impact": "total"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-05-08T23:17:23.691346Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-05-08T23:18:11.328Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "references": [
              {
                "tags": [
                  "exploit"
                ],
                "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-rg3h-x3jw-7jm5"
              }
            ],
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "praisonaiagents \u003c 1.6.9"
                },
                {
                  "status": "affected",
                  "version": "praisonai \u003c 4.6.9"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "PraisonAI is a multi-agent teams system. Prior to praisonai version 4.6.9 and praisonaiagents version 1.6.9, the fix for CVE-2026-40315 added input validation to SQLiteConversationStore only. Nine sibling backends \u2014 MySQL, PostgreSQL, async SQLite/MySQL/PostgreSQL, Turso, SingleStore, Supabase, SurrealDB \u2014 pass table_prefix straight into f-string SQL. Same root cause, same code pattern, same exploitation. 52 unvalidated injection points across the codebase. postgres.py additionally accepts an unvalidated schema parameter used directly in DDL. This issue has been patched in praisonai version 4.6.9 and praisonaiagents version 1.6.9."
            }
          ],
          "metrics": [
            {
              "cvssV3_1": {
                "attackComplexity": "LOW",
                "attackVector": "NETWORK",
                "availabilityImpact": "NONE",
                "baseScore": 8.1,
                "baseSeverity": "HIGH",
                "confidentialityImpact": "HIGH",
                "integrityImpact": "HIGH",
                "privilegesRequired": "LOW",
                "scope": "UNCHANGED",
                "userInteraction": "NONE",
                "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:N",
                "version": "3.1"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-89",
                  "description": "CWE-89: Improper Neutralization of Special Elements used in an SQL Command (\u0027SQL Injection\u0027)",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-05-08T13:19:10.753Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-rg3h-x3jw-7jm5",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-rg3h-x3jw-7jm5"
            }
          ],
          "source": {
            "advisory": "GHSA-rg3h-x3jw-7jm5",
            "discovery": "UNKNOWN"
          },
          "title": "PraisonAI: SQL Injection via unvalidated `table_prefix` in 9 conversation store backends (incomplete fix for CVE-2026-40315)"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-41496",
        "datePublished": "2026-05-08T13:19:10.753Z",
        "dateReserved": "2026-04-20T16:14:19.009Z",
        "dateUpdated": "2026-05-08T23:18:11.328Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }

    CVE-2026-40289 (GCVE-0-2026-40289)

    Vulnerability from cvelistv5 – Published: 2026-04-14 03:05 – Updated: 2026-04-14 20:18
    VLAI
    Title
    PraisonAI Browser Server allows unauthenticated WebSocket clients to hijack connected extension sessions
    Summary
    PraisonAI is a multi-agent teams system. In versions below 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents, the browser bridge (praisonai browser start) is vulnerable to unauthenticated remote session hijacking due to missing authentication and a bypassable origin check on its /ws WebSocket endpoint. The server binds to 0.0.0.0 by default and only validates the Origin header when one is present, meaning any non-browser client that omits the header is accepted without restriction. An unauthenticated network attacker can connect, send a start_session message, and the server will route it to the first idle browser-extension WebSocket (effectively hijacking that session) and then broadcast all resulting automation actions and outputs back to the attacker. This enables unauthorized remote control of connected browser automation sessions, leakage of sensitive page context and automation results, and misuse of model-backed browser actions in any environment where the bridge is network-reachable. This issue has been fixed in versions 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents.
    SSVC
    Exploitation: poc Automatable: yes Technical Impact: total
    CISA Coordinator · CISA-ADP (v2.0.3)
    Decision recorded 2026-04-14 20:18 UTC
    CWE
    • CWE-306 - Missing Authentication for Critical Function
    Assigner
    GitHub_M CNA under the mitre root
    CNA scorecard C 72/100 over 8156 records in the last 180 days details
    References
    Impacted products
    Show details on NVD website

    {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-40289",
                    "options": [
                      {
                        "Exploitation": "poc"
                      },
                      {
                        "Automatable": "yes"
                      },
                      {
                        "Technical Impact": "total"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-04-14T20:18:27.623888Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-04-14T20:18:37.319Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 4.5.139"
                }
              ]
            },
            {
              "product": "praisonaiagents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.140"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "PraisonAI is a multi-agent teams system. In versions below 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents, the browser bridge (praisonai browser start) is vulnerable to unauthenticated remote session hijacking due to missing authentication and a bypassable origin check on its /ws WebSocket endpoint. The server binds to 0.0.0.0 by default and only validates the Origin header when one is present, meaning any non-browser client that omits the header is accepted without restriction. An unauthenticated network attacker can connect, send a start_session message, and the server will route it to the first idle browser-extension WebSocket (effectively hijacking that session) and then broadcast all resulting automation actions and outputs back to the attacker. This enables unauthorized remote control of connected browser automation sessions, leakage of sensitive page context and automation results, and misuse of model-backed browser actions in any environment where the bridge is network-reachable. This issue has been fixed in versions 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents."
            }
          ],
          "metrics": [
            {
              "cvssV3_1": {
                "attackComplexity": "LOW",
                "attackVector": "NETWORK",
                "availabilityImpact": "NONE",
                "baseScore": 9.1,
                "baseSeverity": "CRITICAL",
                "confidentialityImpact": "HIGH",
                "integrityImpact": "HIGH",
                "privilegesRequired": "NONE",
                "scope": "UNCHANGED",
                "userInteraction": "NONE",
                "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N",
                "version": "3.1"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-306",
                  "description": "CWE-306: Missing Authentication for Critical Function",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-04-14T03:05:05.514Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-8x8f-54wf-vv92",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-8x8f-54wf-vv92"
            }
          ],
          "source": {
            "advisory": "GHSA-8x8f-54wf-vv92",
            "discovery": "UNKNOWN"
          },
          "title": "PraisonAI Browser Server allows unauthenticated WebSocket clients to hijack connected extension sessions"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-40289",
        "datePublished": "2026-04-14T03:05:05.514Z",
        "dateReserved": "2026-04-10T20:22:44.035Z",
        "dateUpdated": "2026-04-14T20:18:37.319Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }

    CVE-2026-40288 (GCVE-0-2026-40288)

    Vulnerability from cvelistv5 – Published: 2026-04-14 03:00 – Updated: 2026-04-14 15:57
    VLAI
    Title
    PraisonAI: Critical RCE via `type: job` workflow YAML
    Summary
    PraisonAI is a multi-agent teams system. In versions below 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents, the workflow engine is vulnerable to arbitrary command and code execution through untrusted YAML files. When praisonai workflow run <file.yaml> loads a YAML file with type: job, the JobWorkflowExecutor in job_workflow.py processes steps that support run: (shell commands via subprocess.run()), script: (inline Python via exec()), and python: (arbitrary Python script execution)—all without any validation, sandboxing, or user confirmation. The affected code paths include action_run() in workflow.py and _exec_shell(), _exec_inline_python(), and _exec_python_script() in job_workflow.py. An attacker who can supply or influence a workflow YAML file (particularly in CI pipelines, shared repositories, or multi-tenant deployment environments) can achieve full arbitrary command execution on the host system, compromising the machine and any accessible data or credentials. This issue has been fixed in versions 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents.
    SSVC
    Exploitation: poc Automatable: yes Technical Impact: total
    CISA Coordinator · CISA-ADP (v2.0.3)
    Decision recorded 2026-04-14 15:56 UTC
    CWE
    • CWE-78 - Improper Neutralization of Special Elements used in an OS Command ('OS Command Injection')
    • CWE-94 - Improper Control of Generation of Code ('Code Injection')
    Assigner
    GitHub_M CNA under the mitre root
    CNA scorecard C 72/100 over 8156 records in the last 180 days details
    References
    Impacted products
    Show details on NVD website

    {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-40288",
                    "options": [
                      {
                        "Exploitation": "poc"
                      },
                      {
                        "Automatable": "yes"
                      },
                      {
                        "Technical Impact": "total"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-04-14T15:56:49.344666Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-04-14T15:57:15.812Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 4.5.139"
                }
              ]
            },
            {
              "product": "praisonaiagents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.140"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "PraisonAI is a multi-agent teams system. In versions below 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents, the workflow engine is vulnerable to arbitrary command and code execution through untrusted YAML files. When praisonai workflow run \u003cfile.yaml\u003e loads a YAML file with type: job, the JobWorkflowExecutor in job_workflow.py processes steps that support run: (shell commands via subprocess.run()), script: (inline Python via exec()), and python: (arbitrary Python script execution)\u2014all without any validation, sandboxing, or user confirmation. The affected code paths include action_run() in workflow.py and _exec_shell(), _exec_inline_python(), and _exec_python_script() in job_workflow.py. An attacker who can supply or influence a workflow YAML file (particularly in CI pipelines, shared repositories, or multi-tenant deployment environments) can achieve full arbitrary command execution on the host system, compromising the machine and any accessible data or credentials. This issue has been fixed in versions 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents."
            }
          ],
          "metrics": [
            {
              "cvssV3_1": {
                "attackComplexity": "LOW",
                "attackVector": "NETWORK",
                "availabilityImpact": "HIGH",
                "baseScore": 9.8,
                "baseSeverity": "CRITICAL",
                "confidentialityImpact": "HIGH",
                "integrityImpact": "HIGH",
                "privilegesRequired": "NONE",
                "scope": "UNCHANGED",
                "userInteraction": "NONE",
                "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
                "version": "3.1"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-78",
                  "description": "CWE-78: Improper Neutralization of Special Elements used in an OS Command (\u0027OS Command Injection\u0027)",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            },
            {
              "descriptions": [
                {
                  "cweId": "CWE-94",
                  "description": "CWE-94: Improper Control of Generation of Code (\u0027Code Injection\u0027)",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-04-14T03:01:53.412Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-vc46-vw85-3wvm",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-vc46-vw85-3wvm"
            }
          ],
          "source": {
            "advisory": "GHSA-vc46-vw85-3wvm",
            "discovery": "UNKNOWN"
          },
          "title": "PraisonAI: Critical RCE via `type: job` workflow YAML"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-40288",
        "datePublished": "2026-04-14T03:00:21.547Z",
        "dateReserved": "2026-04-10T20:22:44.035Z",
        "dateUpdated": "2026-04-14T15:57:15.812Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }

    CVE-2026-40287 (GCVE-0-2026-40287)

    Vulnerability from cvelistv5 – Published: 2026-04-14 02:55 – Updated: 2026-04-14 13:23
    VLAI
    Title
    PraisonAI has RCE via Automatic tools.py Import
    Summary
    PraisonAI is a multi-agent teams system. Versions 4.5.138 and below are vulnerable to arbitrary code execution through automatic, unsanitized import of a tools.py file from the current working directory. Components including call.py (import_tools_from_file()), tool_resolver.py (_load_local_tools()), and CLI tool-loading paths blindly import ./tools.py at startup without any validation, sandboxing, or user confirmation. An attacker who can place a malicious tools.py in the directory where PraisonAI is launched (such as through a shared project, cloned repository, or writable workspace) achieves immediate arbitrary Python code execution in the host environment. This compromises the full PraisonAI process, the host system, and any connected data or credentials. This issue has been fixed in version 4.5.139.
    SSVC
    Exploitation: poc Automatable: no Technical Impact: total
    CISA Coordinator · CISA-ADP (v2.0.3)
    Decision recorded 2026-04-14 13:23 UTC
    CWE
    • CWE-94 - Improper Control of Generation of Code ('Code Injection')
    • CWE-426 - Untrusted Search Path
    Assigner
    GitHub_M CNA under the mitre root
    CNA scorecard C 72/100 over 8156 records in the last 180 days details
    References
    Impacted products
    Show details on NVD website

    {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-40287",
                    "options": [
                      {
                        "Exploitation": "poc"
                      },
                      {
                        "Automatable": "no"
                      },
                      {
                        "Technical Impact": "total"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-04-14T13:23:23.221160Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-04-14T13:23:29.807Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "references": [
              {
                "tags": [
                  "exploit"
                ],
                "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-g985-wjh9-qxxc"
              }
            ],
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "PraisonAI",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 4.5.139"
                }
              ]
            },
            {
              "product": "praisonaiagents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.140"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "PraisonAI is a multi-agent teams system. Versions 4.5.138 and below are vulnerable to arbitrary code execution through automatic, unsanitized import of a tools.py file from the current working directory. Components including call.py (import_tools_from_file()), tool_resolver.py (_load_local_tools()), and CLI tool-loading paths blindly import ./tools.py at startup without any validation, sandboxing, or user confirmation. An attacker who can place a malicious tools.py in the directory where PraisonAI is launched (such as through a shared project, cloned repository, or writable workspace) achieves immediate arbitrary Python code execution in the host environment. This compromises the full PraisonAI process, the host system, and any connected data or credentials. This issue has been fixed in version 4.5.139."
            }
          ],
          "metrics": [
            {
              "cvssV3_1": {
                "attackComplexity": "LOW",
                "attackVector": "LOCAL",
                "availabilityImpact": "HIGH",
                "baseScore": 8.4,
                "baseSeverity": "HIGH",
                "confidentialityImpact": "HIGH",
                "integrityImpact": "HIGH",
                "privilegesRequired": "NONE",
                "scope": "UNCHANGED",
                "userInteraction": "NONE",
                "vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
                "version": "3.1"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-94",
                  "description": "CWE-94: Improper Control of Generation of Code (\u0027Code Injection\u0027)",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            },
            {
              "descriptions": [
                {
                  "cweId": "CWE-426",
                  "description": "CWE-426: Untrusted Search Path",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-04-14T02:55:38.270Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-g985-wjh9-qxxc",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-g985-wjh9-qxxc"
            }
          ],
          "source": {
            "advisory": "GHSA-g985-wjh9-qxxc",
            "discovery": "UNKNOWN"
          },
          "title": "PraisonAI has RCE via Automatic tools.py Import"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-40287",
        "datePublished": "2026-04-14T02:55:38.270Z",
        "dateReserved": "2026-04-10T20:22:44.035Z",
        "dateUpdated": "2026-04-14T13:23:29.807Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }

    CVE-2026-40160 (GCVE-0-2026-40160)

    Vulnerability from cvelistv5 – Published: 2026-04-10 16:59 – Updated: 2026-04-13 15:36
    VLAI
    Title
    PraisonAIAgents has SSRF via unvalidated URL in `web_crawl` httpx fallback
    Summary
    PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, web_crawl's httpx fallback path passes user-supplied URLs directly to httpx.AsyncClient.get() with follow_redirects=True and no host validation. An LLM agent tricked into crawling an internal URL can reach cloud metadata endpoints (169.254.169.254), internal services, and localhost. The response content is returned to the agent and may appear in output visible to the attacker. This fallback is the default crawl path on a fresh PraisonAI installation (no Tavily key, no Crawl4AI installed). This vulnerability is fixed in 1.5.128.
    SSVC
    Exploitation: poc Automatable: no Technical Impact: partial
    CISA Coordinator · CISA-ADP (v2.0.3)
    Decision recorded 2026-04-13 15:28 UTC
    CWE
    • CWE-918 - Server-Side Request Forgery (SSRF)
    Assigner
    GitHub_M CNA under the mitre root
    CNA scorecard C 72/100 over 8156 records in the last 180 days details
    References
    Impacted products
    Show details on NVD website

    {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-40160",
                    "options": [
                      {
                        "Exploitation": "poc"
                      },
                      {
                        "Automatable": "no"
                      },
                      {
                        "Technical Impact": "partial"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-04-13T15:28:31.603808Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-04-13T15:36:41.165Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "references": [
              {
                "tags": [
                  "exploit"
                ],
                "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-qq9r-63f6-v542"
              }
            ],
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "PraisonAIAgents",
              "vendor": "MervinPraison",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 1.5.128"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, web_crawl\u0027s httpx fallback path passes user-supplied URLs directly to httpx.AsyncClient.get() with follow_redirects=True and no host validation. An LLM agent tricked into crawling an internal URL can reach cloud metadata endpoints (169.254.169.254), internal services, and localhost. The response content is returned to the agent and may appear in output visible to the attacker. This fallback is the default crawl path on a fresh PraisonAI installation (no Tavily key, no Crawl4AI installed). This vulnerability is fixed in 1.5.128."
            }
          ],
          "metrics": [
            {
              "cvssV4_0": {
                "attackComplexity": "LOW",
                "attackRequirements": "PRESENT",
                "attackVector": "NETWORK",
                "baseScore": 7.1,
                "baseSeverity": "HIGH",
                "privilegesRequired": "NONE",
                "subAvailabilityImpact": "NONE",
                "subConfidentialityImpact": "HIGH",
                "subIntegrityImpact": "LOW",
                "userInteraction": "PASSIVE",
                "vectorString": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:P/VC:H/VI:N/VA:N/SC:H/SI:L/SA:N",
                "version": "4.0",
                "vulnAvailabilityImpact": "NONE",
                "vulnConfidentialityImpact": "HIGH",
                "vulnIntegrityImpact": "NONE"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-918",
                  "description": "CWE-918: Server-Side Request Forgery (SSRF)",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-04-10T16:59:09.542Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-qq9r-63f6-v542",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-qq9r-63f6-v542"
            }
          ],
          "source": {
            "advisory": "GHSA-qq9r-63f6-v542",
            "discovery": "UNKNOWN"
          },
          "title": "PraisonAIAgents has SSRF via unvalidated URL in `web_crawl` httpx fallback"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-40160",
        "datePublished": "2026-04-10T16:59:09.542Z",
        "dateReserved": "2026-04-09T19:31:56.014Z",
        "dateUpdated": "2026-04-13T15:36:41.165Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }