Issue Station
Sign In
Tools hereToolbox

openclaw - 💡(How to fix) Fix Feature: skill quality testing with agent-skill-infra [1 comments, 2 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
openclaw/openclaw#75958•Fetched 2026-05-03 04:43:53
View on GitHub
Comments
1
Participants
2
Timeline
3
Reactions
2
Author
Liber1917Liber1917
Participants
clawsweeper[bot]clawsweeper[bot]Liber1917Liber1917
Timeline (top)
closed ×1commented ×1unsubscribed ×1

agent-skill-infra is a quality testing toolkit for Agent Skills, compatible with OpenClaw's SKILL.md format. Three modules on top of the agentskills.io spec: quality scoring, behavior testing, and version awareness.

Root Cause

agent-skill-infra is a quality testing toolkit for Agent Skills, compatible with OpenClaw's SKILL.md format. Three modules on top of the agentskills.io spec: quality scoring, behavior testing, and version awareness.

Code Example

$ pip install agent-skill-infra
$ skill-quality skills/skill-creator/SKILL.md --gh-models --output json
{"overall_score": 0.85, "dimensions": [...], "findings": ["..."], "improvements": ["..."]}
RAW_BUFFERClick to expand / collapse

Summary

agent-skill-infra is a quality testing toolkit for Agent Skills, compatible with OpenClaw's SKILL.md format. Three modules on top of the agentskills.io spec: quality scoring, behavior testing, and version awareness.

Why OpenClaw

OpenClaw is the largest SKILL.md-native platform with 20+ built-in skills. Each skill lives in openclaw/skills/<name>/SKILL.md — the exact format agent-skill-infra was built to test and score.

What this adds

Quality scoring: 8-dimension semantic evaluation via GitHub Models (free, zero API key). Every score comes with concrete improvement suggestions — not just a number.

$ pip install agent-skill-infra
$ skill-quality skills/skill-creator/SKILL.md --gh-models --output json
{"overall_score": 0.85, "dimensions": [...], "findings": ["..."], "improvements": ["..."]}

Behavior testing: 5 judge types (keyword, schema, LLM, flow, snapshot) for automated regression detection on skill outputs.

Version awareness: Git-based diff, rollback, and baseline comparison — catches drift before users do.

Integration options

LevelEffortWhat it does
CLI opt-inZeroDevs run skill-quality locally on their skills
CI checkLowAdd a GitHub Action to score skills on PR
ClawHub integrationMediumShow quality scores in ClawHub skill listings

Quick facts

  • Python 3.12+, MIT license, PyPI installed
  • 221 tests, v0.3.0
  • Compatible with agentskills.io spec
  • Free tier via GitHub Models (gpt-4o-mini, GITHUB_TOKEN auto-injected)

Would love feedback on whether this type of quality infrastructure would be useful in the OpenClaw ecosystem.

extent analysis

TL;DR

To integrate agent-skill-infra into the OpenClaw ecosystem, consider starting with the CLI opt-in option for local testing, followed by CI check integration for automated scoring on PRs.

Guidance

  • Evaluate the compatibility of agent-skill-infra with the OpenClaw platform by reviewing the agentskills.io spec and SKILL.md format.
  • Assess the effort required for each integration option (CLI opt-in, CI check, ClawHub integration) to determine the best approach.
  • Review the 8-dimension semantic evaluation and behavior testing features to understand how they can benefit the OpenClaw ecosystem.
  • Explore the use of GitHub Models (gpt-4o-mini) and the auto-injected GITHUB_TOKEN for free tier access.

Notes

The issue lacks specific technical details about the integration process, so further clarification may be needed to determine the best course of action.

Recommendation

Apply the CLI opt-in workaround to start testing agent-skill-infra locally, allowing for a low-effort evaluation of its usefulness in the OpenClaw ecosystem.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…
#api#permission error#memory optimization#batch processing#GPU compatibility

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING

openclaw - 💡(How to fix) Fix Feature: skill quality testing with agent-skill-infra [1 comments, 2 participants]