autofix

Use when Qwen Code Autofix runs from GitHub Actions or an operator dry-run to choose an approved issue, implement it, or address review feedback on an existing autofix PR.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

Highly actionable and clearly sequenced with strong validation checkpoints and feedback loops, but it is token-heavy due to repeated verification-command enumeration and is monolithic rather than progressively disclosed into the skill's bundle.

Suggestions

DRY the verification-command catalog: state the trusted command list once in Shared Rules and have each mode reference it ("run the verification commands per Shared Rules") instead of repeating the full build/typecheck/lint/vitest/bundle/generate:settings-schema enumeration three times.

Move the bilingual-output format spec and the detailed per-mode rule catalogs into a reference file under references/ (and link to it), reducing SKILL.md to an overview and improving progressive disclosure.

If scripts/run-agent.mjs is part of this skill's bundle, reference it from the body where relevant; otherwise clarify why it exists so the bundle structure is discoverable from SKILL.md.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and assumes Claude's competence, but the trusted-verification command list (build/typecheck/lint/vitest/bundle/generate:settings-schema) is repeated verbatim across Shared Rules, develop-issue step 6, and address-review, and the bilingual-output rule is restated — tightening would remove redundancy.	2 / 3
Actionability	Fully executable guidance throughout — exact commands, file paths like `<workdir>/decision.json`/`failure.md`/`resolved-comments.txt`, branch name `autofix/issue-<issue>`, a decision.json schema, and Conventional Commit format — copy-paste ready.	3 / 3
Workflow Clarity	develop-issue is a clearly numbered 1–9 sequence with an explicit verification checkpoint, a "re-read the diff as a skeptical reviewer" step, and a failure.md stop rule; address-review has feedback classification, conflict handling, and a one-outcome finish with a run-the-gate-first feedback loop.	3 / 3
Progressive Disclosure	Well sectioned into Shared Rules and three modes, but it is a monolithic ~250-line single file with no offloading to the skill's own bundle (the provided scripts/run-agent.mjs is not referenced from the body), and content that could be split — the bilingual-format spec and per-mode rule catalogs — is inline.	2 / 3
	Total	10 / 12 Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description that names three concrete actions, gives an explicit "Use when" trigger tied to the system's run contexts, and occupies a distinct niche unlikely to conflict with other skills.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "choose an approved issue, implement it, or address review feedback on an existing autofix PR" — rather than vague language, matching the multiple-specific-actions anchor.	3 / 3
Completeness	Explicitly answers both what (choose/implement/address-review) and when via an explicit "Use when..." clause, matching the both-what-and-when-with-explicit-triggers anchor.	3 / 3
Trigger Term Quality	Covers natural trigger contexts for this domain — "Qwen Code Autofix runs from GitHub Actions", "operator dry-run", "approved issue", "review feedback", "autofix PR" — giving good coverage of terms an operator would say.	3 / 3
Distinctiveness Conflict Risk	Scoped to a clear niche — Qwen Code Autofix, GitHub Actions, autofix PR review — with distinct triggers unlikely to collide with other skills.	3 / 3
	Total	12 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: QwenLM/qwen-code
Path: .qwen/skills/autofix/SKILL.md
Commit: 62e009a

Reviewed: 5 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.