defillama-pro-openapi-skill
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches a curated OpenAPI schema from the vendor's repository on GitHub (github.com/holon-run/uxc). This is a legitimate resource used to define the API structure for the uxc tool.
- [CREDENTIALS_UNSAFE]: The skill provides explicit security guidance for managing DefiLlama Pro API keys. It correctly instructs the user to use environment variables (
--secret-env) to prevent keys from appearing in shell history. It also includes specific guardrails warning that keys will appear in URL paths and advises on log rotation and sanitization. - [INDIRECT_PROMPT_INJECTION]: The skill interacts with external data from
pro-api.llama.fi. While this creates an ingestion point for untrusted data, the skill specifies read-only analytics operations and encourages parsing the structured JSON output envelope, which limits the potential for malicious content to influence agent behavior. - [COMMAND_EXECUTION]: Uses the
uxctool to manage bindings and execute API operations. The command structure is well-defined and constrained by the provided OpenAPI schema, minimizing the risk of arbitrary command injection.
Audit Metadata