deepwiki

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly instructs the agent to fetch and read documentation from public third-party sites (e.g., WebFetch https://deepwiki.com/owner/repo and MCP tools like mcp__deepwiki__read_wiki_contents, as well as fallback GitHub API calls to read READMEs) so untrusted user-generated or web-hosted content would be ingested and could influence agent actions.

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

• Potentially malicious external URL detected (high risk: 0.80). This skill calls DeepWiki endpoints at runtime (e.g., https://deepwiki.com/{owner}/{repo} and the MCP endpoints https://mcp.deepwiki.com/mcp and https://mcp.deepwiki.com/sse) to fetch documentation that is injected into the agent's context and used to drive responses, so remote content can directly influence prompts.