git-workflow Eval
=================

Status: failed
Score: 61/100
Risk: high
Decision: do_not_auto_install
Policy: block
Reason: Agent safety gate: This skill should not be selected by an agent without explicit human security review.

Install:
npx skills add huangwb8/skills --skill git-workflow

Required checks:
- PASS Task fit: Task wording matches this skill metadata.
- PASS Install path: Install handoff is available.
- PASS Install command safety: standard package or runtime install path
- WARN Trust score: Potentially useful, but at least one trust signal needs human inspection.
- WARN Audit score: Needs review
- FAIL Agent safety gate: This skill should not be selected by an agent without explicit human security review.
- PASS License clarity: MIT
- FAIL Permission surface: secrets or environment access, shell or command execution

Warnings:
- Trust score: Potentially useful, but at least one trust signal needs human inspection.
- Audit score: Needs review
- High-risk permission hints: Shell or command execution, Secrets or environment access
- Dependency or permission surface needs review
- Permission surface may require sandboxing
- The SKILL.md embeds BenszAPI-specific workspace and bensz-collect-bugs collaboration rules, which instruct the agent to create hidden directories and write outside the repository. This is surprising for a generic git-workflow skill and may cause unwanted side effects.
- The skill lacks explicit setup/prerequisites and safety boundaries, such as guidance about force-pushing to shared branches, handling Git credentials, or what to do when not in a Git repository.
- The PR review response section appears truncated mid-sentence in the provided excerpt; the full SKILL.md should be verified for completeness.
- The description mentions automated workflows, but the skill content focuses on conventions and templates without providing actual automation examples or configurations.
- Low GitHub adoption signal
- Quality score needs review
- Permission surface needs review: secrets or environment access, shell or command execution

Validation plan:
1. Inspect repository, README/SKILL.md, license, and recent commits before production use.
2. Install in an isolated workspace or sandbox with no production secrets available.
3. Run the smallest representative task and record files touched, commands run, network access, and outputs.
4. Compare the selected skill against at least one alternative when the eval status is review or failed.
5. Promote only after the agent reports a successful verification result and unresolved warnings are accepted.

Do not use when:
- teams that need a vendor-supported SLA
- production agents without a repository review
- Low GitHub adoption signal
- The SKILL.md embeds BenszAPI-specific workspace and bensz-collect-bugs collaboration rules, which instruct the agent to create hidden directories and write outside the repository. This is surprising for a generic git-workflow skill and may cause unwanted side effects.
- No OpenAgentSkill engagement data yet
- High-risk permission hints: Shell or command execution, Secrets or environment access
- Dependency or permission surface needs review
- Permission surface may require sandboxing

URLs:
- Skill: https://www.openagentskill.com/skills/huangwb8-git-workflow
- Audit: https://www.openagentskill.com/skills/huangwb8-git-workflow/audit
- JSON: https://www.openagentskill.com/api/agent/evals?slug=huangwb8-git-workflow