langgraph-docs

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.95). At runtime the workflow uses fetch_url to retrieve public web pages from https://docs.langchain.com/llms.txt and then fetches additional documentation URLs from that index, so outsider-authored free text from the public web is ingested into the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 1.00). This skill uses fetch_url at runtime to load remote documentation (starting with https://docs.langchain.com/llms.txt and then the selected URLs from that index) and injects that fetched content into the agent context to control prompts and responses.