release-publisher

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes git and gh (GitHub CLI) to manage tags, commits, and releases. It also executes a project-local PowerShell script .\tests\scripts\run-tests.ps1 to perform end-to-end validation before publishing. These commands are integral to the skill's purpose and follow best practices for release automation.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from git log and GitHub release metadata to generate release notes and social media copy. However, the instructions implement several safety measures: they require summarizing and categorizing commits into specific functional modules rather than raw inclusion, and explicitly instruct the agent to ignore specific CI-related trigger words. This sanitization minimizes the risk of embedded instructions in commit messages influencing the agent's behavior.
  • [SAFE]: The skill adheres to the principle of least privilege by checking authentication status via gh auth status and requesting user confirmation before performing the final GitHub Release creation. It also protects existing work by verifying the clean state of the git repository before proceeding with automated updates.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 02:52 PM
Security Audit — agent-trust-hub — release-publisher