context7
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill imports text documentation from an external API, introducing a potential indirect prompt injection surface if the external source contains manipulative language or instructions.
- Ingestion points: In
scripts/context7.sh, thefetch_docsfunction performs acurlrequest to retrieve text data fromhttps://context7.com/api/v2/docs/. - Boundary markers: No specific encapsulation or delimiters are placed around the fetched output to distinguish it from system instructions.
- Capability inventory: The skill is restricted to network fetch queries via
curland text parsing viajqinsidescripts/context7.sh. There are no system modification or file-writing capabilities. - Sanitization: Content is passed directly from the API response to the standard output without sanitization or structural validation.
Audit Metadata