defillama-pro-openapi-skill
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill provides defensive guidance for managing DefiLlama Pro API keys, recommending environment variables via
--secret-envand instructing users to sanitize local daemon logs where keys might appear in URL paths. - [COMMAND_EXECUTION]: The skill utilizes the
uxctool to interact withpro-api.llama.fi. These operations are limited to the scope of the curated OpenAPI schema provided by the vendor. - [SAFE]: While the skill ingests external data from the DefiLlama API, the risk of indirect prompt injection is limited by its read-only focus and the requirement for structured JSON output.
- Ingestion points: API responses from
pro-api.llama.fi(SKILL.md). - Boundary markers: The skill mandates the use of JSON output for automation, avoiding text-based interpretations (SKILL.md).
- Capability inventory: Interaction with the DefiLlama Pro API host as defined in the OpenAPI schema (references/defillama-pro.openapi.json).
- Sanitization: The skill relies on the structured nature of the JSON response format.
Audit Metadata