answeroverflow

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

• Third-party content exposure detected (high risk: 0.90). This skill (SKILL.md) explicitly directs the agent to search and fetch public Answer Overflow threads—indexed Discord community conversations—via web_search, web_fetch, and the MCP server (see "Fetching Thread Content" and "MCP Server"), meaning the agent will read untrusted user-generated content that can influence its actions.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill explicitly fetches external markdown threads at runtime (e.g., https://www.answeroverflow.com/m/ and via the MCP endpoint https://www.answeroverflow.com/mcp) which are injected into the agent's context to produce answers, so external content can directly influence prompts/instructions.