deepwiki

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly fetches and reads documentation from public GitHub repositories via the DeepWiki MCP (see SKILL.md and reference.md which list the mcp.deepwiki.com server and the read_wiki_contents/read_wiki_structure/ask_question calls), so it ingests untrusted, user-generated third-party content that the agent is expected to interpret and act on.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill calls the remote DeepWiki MCP endpoint https://mcp.deepwiki.com/mcp at runtime to fetch repository documentation that is injected into the agent’s context to drive prompts/answers, so this external URL directly controls agent prompts and is a required dependency.