recon-hvac

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly instructs retrieving and printing debug logs, contact data, and searching for/exposing embedded API keys (e.g., Nest/Google Maps) which requires the agent to output sensitive secrets and PII verbatim.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This skill is an explicit targeted recon guide that instructs how to enumerate sites, harvest debug logs and contact form data, test CORS with a malicious origin, enumerate IDORs, and locate API keys — all facilitating data exfiltration and abuse.

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md’s required workflow uses runtime-fetched public web content (e.g., curl to https://$TARGET/wp-content/debug.log, .../wp-content/uploads/..., and other site pages) and then parses/prints it (PII extraction, directory listing, plugin readme.txt), which becomes readable free text in the agent’s context.