crawler

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly instructs capturing and documenting cookies, tokens, and headers and to write raw auth-signal files, which requires the agent to include secret values verbatim in generated artifacts (exfiltration risk).

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The skill content includes explicit, repeated instructions to capture cookies/tokens/Authorization headers and to bypass anti-bot protections (TLS/JA3 impersonation, browser binary patches, proxy rotation, CAPTCHA solver services), which enable credential theft and stealthy evasion—high risk for abuse.

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md’s required runtime workflow performs Playwright-driven page exploration and then captures network/auth evidence via chrome-devtools-mcp/CDP (e.g., list_network_requests/get_network_request or Network.*/Runtime.evaluate) from the target website’s live responses, which are outsider-authored free text when the target is user-controlled.