babysit
Pass
Audited by Gen Agent Trust Hub on Jul 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill facilitates a standard software development workflow for resolving PR conflicts and CI logs without malicious intent.
- [PROMPT_INJECTION]: Indirect Prompt Injection Risk Assessment:
- Ingestion points: The skill ingests untrusted data from GitHub pull request comments, PR descriptions, and CI logs via
gh pr viewand reading unresolved threads (documented inSKILL.mdunder 'The loop' and 'Comment triage'). - Boundary markers: Absent. The skill does not explicitly instruct the agent to ignore instructions embedded in the external content, though it does tell the agent to "validate each; act only on real issues."
- Capability inventory: The skill has the capability to write to the repository (
git push,git merge,git rebase) as described in the conflict resolution and CI fixing sections. - Sanitization: Absent. The skill processes the raw text of comments and logs. The instruction to "validate carefully" serves as a behavioral guideline rather than a technical sanitization step.
Audit Metadata