langgraph-orchestration

Fail

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The provided code for the calculate tool uses the eval() function to process the expression argument. In an AI agent environment, this is a critical risk because the LLM—which generates the tool parameters—can be manipulated via prompt injection to execute arbitrary Python code.\n
  • Evidence: return eval(expression) in the calculate tool function in skill.md.\n- [COMMAND_EXECUTION]: Exploiting the eval() vulnerability allows an attacker to execute operating system commands (e.g., using __import__('os').system()), which could lead to full system compromise or data exfiltration.\n- [EXTERNAL_DOWNLOADS]: The skill recommends installing langgraph and langchain-openai via pip. These are well-known and trusted packages from established organizations.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Jul 30, 2026, 04:00 PM
Security Audit — agent-trust-hub — langgraph-orchestration