context7-mcp

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). This skill explicitly instructs the agent to pass the user's "full question" verbatim into external calls (resolve-library-id and query-docs), which would cause any user-supplied secrets (API keys, tokens, passwords) included in the question to be embedded in requests/outputs and thus exposed.

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md instructs the runtime to call Context7 (resolve-library-id → query-docs) to fetch “current documentation,” which implies pulling public/outsider-authored web documentation text into the agent’s LLM context.