deepagents-typescript-quickstart

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

• Third-party content exposure detected (high risk: 0.70). The required workflow instructs fetching the live docs from the public URL (https://docs.langchain.com/...) at runtime and then using that retrieved page text to shape the agent, which is outsider-authored free text entering the LLM context indirectly via the “Docs MCP or HTTP” path.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs the implementer/agent to "Fetch that page" and follow the live docs at https://docs.langchain.com/oss/javascript/deepagents/quickstart, which is a runtime fetch of external content that will directly control the agent's prompt/implementation shape and is required for the skill to run.