release

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • Execution of repository-specific scripts: The skill executes scripts/bump_version.sh located within the repository. This is a common method for handling versioning logic tailored to a specific project. Users should ensure the script's content is trusted before execution.\n- Command execution with development tools: The skill utilizes standard tools such as git and gh (GitHub CLI) to perform release tasks like tagging, pushing code, and merging pull requests. These are powerful capabilities necessary for the skill's primary purpose of release management.\n- Potential surface for indirect prompt injection: The skill processes external data from the GitHub API (gh repo view, gh pr view) and the PyPI JSON API to establish release state. Ingestion points include command outputs and external registry metadata. While no specific boundary markers or sanitization steps are defined in the instructions, the skill includes logic to verify specific commit hashes and version strings, which helps ensure the integrity of the release process. Capabilities linked to this data ingestion include git push and gh pr merge.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 04:26 AM
Security Audit — agent-trust-hub — release