hardcoded-credential-hunt

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly extracts secret values from responses and prints/inserts them verbatim into outputs and HTTP requests (including using found passwords in curl commands), which requires the LLM to handle secret values directly and risks exfiltration.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document contains explicit, automated instructions to locate and extract hardcoded secrets from remote systems and to test them against authentication endpoints, facilitating credential theft and unauthorized access.

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

• Third-party content exposure detected (medium risk: 0.30). The skill reads and analyzes attacker-influenced runtime free text from the target’s untrusted HTML/JSON/JS responses fetched via curl (e.g., HTML form bodies, /api/config JSON, debug/error page HTML, and auth endpoint responses), so outsider-authored text can be supplied by the target/source those HTTP calls return.