recon-churches

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill includes commands that embed secrets verbatim (a Cookie header variable in a curl request) and a plaintext password inside an XMLRPC request, which would require an agent to insert secret values directly into generated commands/requests.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document is explicit offensive tooling for targeting church websites, describing reconnaissance and direct attack chains (CORS credential reflection, XMLRPC brute-force, debug/wp-config exfiltration, credential harvesting) and is clearly malicious in intent.

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md’s required workflow fetches public web content at runtime (e.g., curl to crt.sh in Phase 1 and to arbitrary target URLs in later phases), and those fetched responses can include outsider-authored free text that may be ingested by the agent/LLM via subsequent parsing/printing (e.g., JSON piped into Python/jq/head).