exa-code-search

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

• Third-party content exposure detected (high risk: 0.85). This skill calls the MCP tool exa-code__get_code_context_exa to fetch code context from Exa (public web/code search results), so outsider-authored free text/code snippets can be ingested into the agent’s LLM context via the tool’s returned readable text.