documentation-lookup
Pass
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill is designed to send user-provided queries to the Context7 MCP server. While it explicitly instructs the agent to redact sensitive data like API keys and tokens before transmission, this still constitutes a flow of user data to an external third-party service.
- [PROMPT_INJECTION]: The skill ingests external content from documentation via the query-docs tool. This creates a vulnerability to indirect prompt injection, where compromised or malicious documentation could influence the agent's behavior. 1. Ingestion points: Results returned by the query-docs MCP tool. 2. Boundary markers: The skill does not define specific markers or instructions to isolate the documentation content from the agent's core instructions. 3. Capability inventory: The agent processes documentation to generate code examples and configuration steps for the user. 4. Sanitization: No inbound sanitization or verification of the documentation content is performed.
Audit Metadata