self-healing-ci

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

• Third-party content exposure detected (high risk: 0.85). The required runtime path for this skill is to read PR diff, CI logs, and existing .learnings/HEALS.md (via GitHub Actions/gh-aw tooling), and CI logs and PR content are outsider-authored free text that the agent ingests into its LLM context for diagnosis.

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

• Potentially malicious external URL detected (high risk: 0.70). The workflows run the external GitHub Action githubnext/gh-aw (https://github.com/githubnext/gh-aw) at runtime to fetch and execute the self-healing skill, which runs remote code that controls agent behavior/instructions.