context7-mcp
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content fetched from external sources via the
query-docstool, which is then incorporated into the agent's response context. This design is susceptible to indirect prompt injection if the source documentation contains adversarial content. - Ingestion points: Data returned by the
query-docstool referenced inSKILL.md. - Boundary markers: The skill lacks instructions for the agent to use delimiters or ignore potential instructions within the fetched documentation content.
- Capability inventory: The skill is designed to resolve library identifiers and query documentation using specified MCP tools to inform agent responses.
- Sanitization: No sanitization, filtering, or validation steps for the external documentation content are described in the skill instructions.
Audit Metadata