crypto-market

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt embeds an explicit API key header (AUTH=(-H "X-API-Key: 123456" ...)) and uses it in curl examples, which requires the LLM to emit the secret header/value verbatim in generated commands/requests.

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly instructs the agent to fetch and use external market and news data (e.g., GET /api/v2/crypto/news/list, /news/detail, /news/search and other realtime endpoints on the external BASE API) as part of its required workflow in SKILL.md, exposing it to untrusted third-party (public news/user-generated) content that the agent must read and that could materially influence actions or decisions.