langchain-orchestration

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

• Third-party content exposure detected (high risk: 0.90). The skill's examples (e.g., EXAMPLES.md Agent examples like "web_search"/"Search the web" and the Zero-Shot React example that loads the "serpapi" tool, plus RAG chains that retrieve external documents) explicitly show the agent fetching and ingesting open web / third‑party search results which are untrusted and then used as context for LLM reasoning, exposing it to indirect prompt injection.