lov-release-via-cicd

Warn

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Automated elevation of GitHub repository security settings. The skill uses gh api to set default_workflow_permissions to write and enables PR approval permissions for the repository.
  • [COMMAND_EXECUTION]: Extensive automation of git and GitHub CLI operations including git add, git commit, git push, and gh workflow run. These automate the full release cycle, including the execution of remote CI/CD pipelines.
  • [EXTERNAL_DOWNLOADS]: The skill provides workflow templates that rely on several third-party GitHub Actions such as softprops/action-gh-release, changesets/action, and tauri-apps/tauri-action for handling releases and build environments.
  • [COMMAND_EXECUTION]: Includes instructions and templates requiring elevated system privileges, such as sudo apt-get for dependency installation and sudo xattr for modifying macOS file quarantine attributes.
  • [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection via project metadata processing:
  • Ingestion points: Reads CHANGELOG.md, branch names, and git commit messages from the repository environment.
  • Boundary markers: Uses awk to extract version-specific sections based on header patterns, but lacks explicit markers to prevent following embedded instructions.
  • Capability inventory: Wide range of capabilities including file system writes, git operations, and repository management via gh CLI.
  • Sanitization: Relies on basic text splitting (regex/awk) without rigorous validation of external content.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 24, 2026, 12:55 AM
Security Audit — agent-trust-hub — lov-release-via-cicd