release
Pass
Audited by Gen Agent Trust Hub on Apr 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Executes shell commands and a release script (
./scripts/release.sh) to automate software versioning, crate publishing, and gateway updates via SSH. These actions are aligned with the skill's stated purpose. - [EXTERNAL_DOWNLOADS]: Fetches and verifies resources from well-known developer services like GitHub and crates.io.
- [PROMPT_INJECTION]: Processes git commit messages into the agent's context for changelog generation, creating a surface for indirect prompt injection.
- Ingestion points: Commit messages retrieved via
git loginSKILL.md. - Boundary markers: No delimiters or warnings are used to separate commit text from agent instructions.
- Capability inventory: Includes execution of project scripts, GitHub API interaction, and remote gateway management.
- Sanitization: No sanitization of the untrusted commit data is specified.
Audit Metadata