alterlab-link-health

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

• Third-party content exposure detected (high risk: 0.80). This skill's playbooks and references explicitly instruct agents to clone repos and fetch public CI logs and to probe and parse arbitrary external URLs/GitHub/PyPI pages (e.g., playbooks/full-audit.md pre-flight: gh repo clone and gh run view --log-failed; references/tier3-substitution.md and playbooks/post-merge.md: curl -sSI -L, gh api repos/...) and then act (substitute/exclude/commit) based on those results, so it plainly consumes untrusted third-party content that can influence actions.