langchain-framework

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

• Third-party content exposure detected (high risk: 0.90). The skill's SKILL.md explicitly shows loading public web content (Document Loading: WebBaseLoader("https://example.com")) and using web/search tools (Agents and Tools: DuckDuckGoSearchRun, WikipediaQueryRun) whose retrieved documents are fed into RAG chains as context, meaning untrusted third‑party pages can be read and can materially influence agent behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). Flagging hub.pull("hwchase17/react") because the code calls hub.pull at runtime to fetch a remote prompt/template (from the LangChain Hub) which is then used directly as the agent's prompt and therefore external content controls agent instructions.