rc-skill-updater

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

• Third-party content exposure detected (high risk: 0.95). Outsider-authored free text from public GitHub repos is fetched at runtime via gh api --hostname github.com ... --jq .content in scripts/check.py (and scripts/apply.py), then decoded from base64 and stored in report.json/upContent (LLM-readable prose).

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

• Potentially malicious external URL detected (high risk: 0.90). The scripts call the GitHub Contents API (via gh api --hostname github.com) at runtime to fetch SKILL.md and related files from the upstream repos (e.g. https://github.com/mattpocock/skills and https://github.com/awesome-skills/code-review-skill — API paths like repos/mattpocock/skills/contents/... and repos/awesome-skills/code-review-skill/contents/...) and those fetched SKILL.md files directly control agent prompts/instructions, so this is a runtime external dependency that can change agent behavior.