gh-fix-ci
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The bundled script
scripts/inspect_pr_checks.pyexecutes system commands includingghandgitusing thesubprocessmodule to retrieve pull request status, checks, and logs. While the command arguments are mostly structured, this provides a functional surface for system interaction. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from GitHub Actions logs, which could be manipulated by external contributors.
- Ingestion points: CI logs are fetched from GitHub via
gh run viewand the GitHub API inscripts/inspect_pr_checks.pyand provided to the agent context. - Boundary markers: The skill does not implement delimiters or specific instructions to treat log snippets as untrusted data, increasing the risk of the agent following instructions embedded in logs.
- Capability inventory: The skill possesses high-impact capabilities, including the ability to propose and implement code changes in the repository and execute further CLI commands.
- Sanitization: Log content is processed and summarized without sanitization or filtering for malicious instruction patterns.
Audit Metadata