documentation-lookup
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external documentation which acts as an ingestion point for untrusted data. 1. Ingestion points: The output of the
query-docstool in SKILL.md. 2. Boundary markers: The instructions lack explicit delimiters or warnings to ignore instructions embedded in the fetched documentation. 3. Capability inventory: The skill allows the agent to generate code and provide technical answers based on external content (SKILL.md). 4. Sanitization: While the skill advises redacting user secrets before sending queries, it does not include logic for sanitizing or validating the content received from the Context7 MCP. - [DATA_EXFILTRATION]: The skill sends user-provided queries to the external Context7 MCP service.
- Network operations: The
resolve-library-idandquery-docstools transmit user input to an external documentation provider. - Mitigation: The instructions include a "No sensitive data" best practice requiring the redaction of secrets before transmission, which reduces the risk of credential exposure.
Audit Metadata