source-leak-hunt

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). Insecure: Step 2 explicitly greps leaked files for DB_/PASSWORD/_KEY/TOKEN patterns and prints/saves those credential lines and raw leaked file contents, which requires the agent to handle and output secret values verbatim.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document provides automated mass-scanning and credential-extraction tooling explicitly aimed at locating and harvesting sensitive files and secrets across many domains, facilitating large-scale data exfiltration and credential theft.

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md describes runtime ingestion of outsider-authored free text by fetching arbitrary paths from user-provided target URLs via curl (e.g., /robots.txt, /.env, /.git/config, /debug.log) and then reading the response body with head/grep for content-verification.