babysit
Pass
Audited by Gen Agent Trust Hub on Aug 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses git and gh (GitHub CLI) to perform automated tasks such as empty commits, pushing branches, and rerunning CI jobs. These are standard operations for the described purpose.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through its data ingestion surface. 1. Ingestion points: Reads PR state and review threads via gh commands in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Includes git push and gh run rerun. 4. Sanitization: No evidence of sanitization for the ingested PR data. This represents a functional requirement rather than a malicious intent.
Audit Metadata