release-skills
Warn
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill implements a 'Project Hook Contract' that executes arbitrary shell commands specified in the
.releaserc.ymlconfiguration file (specifically theprepare_artifactandpublish_artifacthooks). This allows for dynamic command execution controlled by files within the repository. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing untrusted data from git commit messages and GitHub PR metadata to generate changelog content.
- Ingestion points: Git log output and GitHub CLI (
gh pr view) command output in Step 4. - Boundary markers: None specified to separate untrusted commit/PR data from agent instructions during generation.
- Capability inventory: The skill has the capability to write to the file system (changelogs, version files) and execute git commands (commit, tag, push).
- Sanitization: No explicit sanitization or validation is mentioned for the external content before it is interpolated into the generated changelog text.
Audit Metadata