AutoGen executes generated code on the host

HighAV-FW-005Runtime safetyFramework pack

What it finds

AutoGen is configured to run model-generated code directly on the local machine.

Generated code is untrusted. Running it on the host gives it the same access as your application — files, network, and credentials.

What an attacker could do: Any code the model produces, including attacker-steered code, would run on the host.

Use DockerCommandLineCodeExecutor (or another isolated executor) with no secrets mounted and network access restricted.

Pro shows the corrected code for your exact file, step by step, and re-scans to confirm the fix. See plans

Finding code: AUTOGEN_LOCAL_CODE_EXECUTION
How it detects: Looks for a risky pattern, and stays quiet when a matching safeguard is present.
Confidence: Definite: the risky setting is written plainly in the file.
Scans: Agent source using the named framework (LangChain, CrewAI, AutoGen, OpenAI Agents SDK, Claude Agent SDK, computer-use tools)
OWASP LLM Top 10: LLM08 - Excessive Agency, LLM02 - Insecure Output Handling
NIST AI RMF: MG-2.2
SOC 2: CC6.8, CC7.1

Paste your agent code or MCP config. The free scan runs all 124 checks in about 10 seconds — no account needed.