web2-recon

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). High-risk offensive recon content: the skill explicitly automates discovery and collection of exposed credentials/secrets (JS/Repo scanning, GitHub dorks, trufflehog/SecretFinder usage) and includes takeover-detection guidance that can be abused to steal credentials or gain unauthorized access.

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 (the “STANDARD RECON PIPELINE” and “CONTINUOUS MONITORING” sections) ingests outsider-authored free text by crawling and scanning web content from a target plus reading GitHub commit metadata/content from external repos (e.g., commit file lists), which is not first-party trusted.