langgraph-multiagent

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill defines a runtime tool fetch_documentation that calls urllib.request.urlopen on allowlisted external docs (e.g., https://langchain-ai.github.io/langgraph/ and https://docs.langchain.com/oss/python/) and the agent's system_prompt explicitly requires calling that tool before answering, so fetched remote content can directly control prompts/instructions at runtime.