github-release

Pass

Audited by Gen Agent Trust Hub on Sep 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill makes extensive use of the GitHub CLI (gh) and git to manage the release lifecycle, including checking release status, tagging commits, and updating release notes.
  • [EXTERNAL_DOWNLOADS]: The validate-reusable-workflows.sh script performs HTTP HEAD requests to raw.githubusercontent.com to verify that pinned reusable workflows still exist, preventing release failures due to upstream changes.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the environment, such as commit messages and pull request descriptions, to generate narrative release notes. While this represents a potential injection surface, the skill provides specific instructions for human review and narrative overhauling to mitigate risks.
  • [SAFE]: The skill implements custom security guards (guard-gh-release.py and guard-lightweight-tag.py) that actively block the agent from executing harmful commands like gh release delete or creating lightweight tags, which would permanently burn tag names in GitHub repositories.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 24, 2026, 05:41 AM
Security Audit — agent-trust-hub — github-release