chainlink-data-feeds-skill
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch documentation from external URLs using tools like WebFetch or curl when local references are insufficient. This is intended for retrieving official technical documentation and is standard for developer utility skills.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external documentation sources, which introduces an inherent surface for indirect prompt injection if the source content were compromised.\n
- Ingestion points: Documentation is retrieved from external URLs into the agent's active context via WebFetch or curl.\n
- Boundary markers: The instructions do not specify explicit delimiters or warnings for the agent to ignore instructions within the fetched documentation content.\n
- Capability inventory: The skill is configured with access to
Bash,Write,Edit, andReadtools, which are necessary for its primary function but could be targeted by injection attacks.\n - Sanitization: No explicit sanitization or validation of the documentation content is performed beyond the agent's internal processing.
Audit Metadata