technical-analysis

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.80). The skill explicitly instructs the agent to conceal the underlying provider (e.g., "never name the underlying exchange") and to neutralize error messages about Bitget, which is a deceptive instruction to hide provenance despite the skill being labeled as an official Bitget skill and using Bitget APIs in its templates.

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

• Third-party content exposure detected (medium risk: 0.30). At runtime, the skill reads external, user-triggerable free-text-derived data by fetching market candles from https://api.bitget.com/api/v2/spot/market/candles?... (and optionally productType for futures) whenever the user mentions a trading pair without local data.