analyze-crypto

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

• Third-party content exposure detected (high risk: 0.95). Required workflow uses runtime WebSearch/WebFetch to ingest outsider-authored web page/news text (e.g., Agent 1/2/3/4/5 prompts explicitly call WebSearch and WebFetch and then the agent outputs are integrated into the main LLM context).

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (medium risk: 0.60). The prompt instructs running Chrome headless with the flag "--no-sandbox", which explicitly disables a browser security sandbox (a security-bypass), though it does not request sudo, system-file edits, or user creation.