coingecko

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). At runtime, the skill calls CoinGecko API endpoints (e.g., cg_coin_data/cg_coin_by_contract/cg_exchange/cg_nft) via proxied_get, and the returned JSON includes free-text fields like description and links that are inserted into the agent’s LLM context as tool output; these are outsider-authored content from a third-party service.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)