OpenAgentSkill
GitHub200+

Pre-install eval

nature-polishing Eval report.

A machine-readable install decision for agents: task fit, Trust Score, Audit Score, install safety, permission surface, and a concrete validation plan before this skill touches a workspace.

Needs reviewMedium riskREVIEW Policy
81
Eval report
78
Trust
85
Audit
69
Agent safety

manual review

Review the audit page, then allow agent install in a sandboxed workflow.

Required gates

Checks an agent must pass before install

Open JSON

Task fit

84

pass

Task wording matches this skill metadata.

  • Evaluate nature-polishing before installing it in an agent workflow
  • automation
  • Video creation workflows; Claude Code teams; teams that value GitHub adoption signals

Install path

92

pass

Install handoff is available.

  • npx skills add Yuan1z0825/nature-skills --skill nature-polishing

Install command safety

92

pass

standard package or runtime install path

  • npx skills add Yuan1z0825/nature-skills --skill nature-polishing

Trust score

78

warn

Good trust signals with a few areas worth checking before rollout.

  • Strong shortlist
  • 39K GitHub stars
  • Apache-2.0

Audit score

85

pass

Safe to try

  • The SKILL.md router does not explicitly mention how to route LaTeX/layout requests to references/latex-layout.md, even though the description claims that coverage.

Agent safety gate

69

warn

Good audit and safety signals with no high-risk permission hints in public metadata.

  • Review the audit page, then allow agent install in a sandboxed workflow.
  • Safe-to-try audit

License clarity

86

pass

Apache-2.0

  • Apache-2.0

Permission surface

86

pass

filesystem or document access

  • Network access: medium
  • Filesystem access: medium

Validation plan

What the agent should do next

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

Do not use when

Conditions that require another skill

  • teams that need a vendor-supported SLA
  • production agents without a repository review
  • The SKILL.md router does not explicitly mention how to route LaTeX/layout requests to references/latex-layout.md, even though the description claims that coverage.
  • The manifest always_load references ../nature-shared/core/* files outside the skill directory; if those shared files are not packaged with the skill, the skill will fail at runtime.
  • The provided SKILL.md excerpt is truncated; ensure the full file is present and that all referenced fragments exist.
  • Production credentials, payments, or irreversible account changes without explicit human review

Supporting checks

Trust signals behind the decision

README/SKILL.md completeness

pass

86

Metadata includes enough usage and workflow context

Recent maintenance

pass

88

2mo since push

Alternatives available

pass

82

Alternative skills are available for comparison.