Skills
Navigate
skills/cryskai/crysk/pr-hero-pr/Gen Agent Trust Hub

pr-hero-pr

Pass

Audited by Gen Agent Trust Hub on Oct 27, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npx -y pr-hero to download and execute the pr-hero package from the NPM registry. This is a functional requirement for the skill's automated review features.
  • [COMMAND_EXECUTION]: Executes several shell commands including git push, gh repo view, and gh pr create to manage branches and pull requests. These are standard operations for the skill's intended purpose.
  • [INDIRECT_PROMPT_INJECTION]: The skill reads data from repository files such as AGENTS.md, CLAUDE.md, and CONTRIBUTING.md to configure PR parameters and target branches. This pattern allows untrusted repository content to influence agent instructions.
  • Ingestion points: .github/pull_request_template.md, AGENTS.md, CONTRIBUTING.md, and CLAUDE.md (via Step 1 and Step 2 in SKILL.md).
  • Boundary markers: None identified; instructions found in these files are followed directly for branch resolution and template structure.
  • Capability inventory: Subprocess execution via shell (git, gh, npx).
  • Sanitization: None; external file content is used to build command arguments and PR bodies without explicit escaping or validation described.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 27, 2026, 10:56 PM