gentle-ai-comment-writer
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to use the
ghCLI to inspect pull request data, including the title and body. This data is external and untrusted, creating a surface where a malicious PR author could embed instructions to manipulate the agent's response. - Ingestion points: Data fetched from
gh pr viewas shown in the Commands section ofSKILL.md. - Boundary markers: Absent. The instructions do not specify delimiters or warnings to ignore instructions found within the PR text.
- Capability inventory: The skill utilizes command execution via the
ghtool. - Sanitization: None provided in the instructions.
- [COMMAND_EXECUTION]: The skill includes a shell command for the GitHub CLI (
gh pr view) to retrieve PR metadata. This is a standard development workflow tool used here for its intended purpose.
Audit Metadata