skill-miner

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

• Third-party content exposure detected (medium risk: 0.30). The workflow reads external agent session logs and candidate skill definitions from GitHub via curl (scripts/match.py, scripts/score.py), which can contain outsider-authored free text.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (medium risk: 0.30). The script scripts/score.py uses curl to fetch raw SKILL.md files from raw.githubusercontent.com during runtime, representing an external repository-based dependency with common/reputable unofficial provenance.