planning

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

• Third-party content exposure detected (high risk: 0.90). The skill's Research phase explicitly requires using Tavily web search and tavily-extract to fetch and ingest content from external URLs (see SKILL.md Phase 0: RESEARCH and references/mcp-usage.md "Tavily — Web Research"), so the agent will read untrusted third‑party web content that can materially influence planning and subsequent tool actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs runtime use of tavily-extract to fetch and inject external documentation (example URL: "https://neon.tech/docs/..."), which would load remote content into the agent's context and thereby directly influence prompts/instructions.