coingecko-openapi-skill
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches a curated OpenAPI specification from a well-known service (GitHub) under the vendor's official repository.
- [COMMAND_EXECUTION]: Utilizes the
uxccommand-line utility to interact with external APIs. All commands are scoped to read-only market data operations defined in the linked schema. - [INDIRECT_PROMPT_INJECTION]: The skill processes data from external API providers, which represents an inherent surface for indirect prompt injection.
- Ingestion points: Market data and token metadata retrieved from CoinGecko and GeckoTerminal APIs.
- Boundary markers: Instructions explicitly require using the JSON output envelope and parsing specific stable fields such as
data,kind, anderror. - Capability inventory: The skill is limited to performing read-only HTTP GET requests through the
uxctool. - Sanitization: Guardrails discourage the use of raw text output, favoring structured JSON parsing to mitigate potential injection within the data stream.
Audit Metadata