babysit
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute GitHub CLI (
gh) commands to monitor PR status and CI checks, which is a legitimate and expected use of development tools for this task. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources that could contain malicious instructions designed to influence the agent's behavior during log analysis.
- Ingestion points: Data enters the context via the output of
gh pr view,gh pr checks, andgh run view --log-failedas defined inSKILL.md. - Boundary markers: The instructions do not utilize specific delimiters or provide guidance to the agent to treat output as untrusted text rather than instructions.
- Capability inventory: The skill relies on the agent's ability to execute shell commands (
gh) to interact with the environment. - Sanitization: There is no mention of sanitizing or filtering the content of external workflow logs before the agent processes them.
Audit Metadata