merge-gate

合入 main 的完整流程：门禁检查 → PR → remote review → Feature Doc Truth 核对(pre-merge) → squash merge → 记录已合入(post-merge) → 清理。 Use when: reviewer 放行后准备合入、开 PR、触发remote review、准备 merge。 Not for: 开发中、review 未通过、自检未完成。 Output: PR merged + worktree cleaned。

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%Weight 40%Scale 1-3

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

The body is highly actionable and workflow-clear, with strong sequencing and validation gates for a complex merge process. Its main weakness is conciseness and progressive disclosure: it is a dense monolith whose detailed protocol tables and policy rationale belong in reference files that are not provided.

Suggestions

Move the 封板协议, Review Provenance Matrix, and Evidence Manifest field-definition tables into separate reference files (e.g. refs/review-provenance.md, refs/evidence-manifest.md) and replace them in SKILL.md with concise summaries plus links.

Tighten repeated content — the merge procedure is described both as inline bash in '合入方式' and again in Step 7 / Step 8 — to reduce the ~660-line footprint.

Provide the referenced bundle files (refs/pr-template.md, refs/pr-signals.md, scripts/check-feature-truth.mjs, etc.) or convert their inline callouts to clearly-resolved one-level-deep references so navigation actually lands on content.

Dimension	Reasoning	Score
Conciseness	The body is a ~660-line, 40KB monolith with dense prose, repeated explanations of the merge steps, and lengthy inline policy detail (封板协议, Review Provenance Matrix, Evidence Manifest field tables); while mostly accurate and domain-specific, much of it could be tightened or moved out of SKILL.md.	2 / 3
Actionability	Provides concrete, copy-paste-ready bash commands with placeholders (e.g. "gh pr merge {PR_NUMBER} --squash --delete-branch", "pnpm gate", the E1–E5 evidence checks) rather than abstract direction.	3 / 3
Workflow Clarity	Steps are explicitly sequenced (Step 0 → 0.5 → 5 → 6 → 6.5 → 6.8 → 6.9 → 7 → 7.5a/b → 7.6 → 8 → 8.5) with hard validation checkpoints (E1–E5 BLOCKED, fail-closed dirty-tree checks, check-feature-truth) and feedback loops (re-trigger review, re-validate).	3 / 3
Progressive Disclosure	Sections are well-organized and inline references are signaled ("详见 refs/pr-signals.md", "refs/pr-template.md"), but no bundle files exist and the heavy policy/protocol content that should live in separate reference files is all inline in SKILL.md, leaving it largely monolithic.	2 / 3
	Total	10 / 12 Passed

Description

100%Weight 40%Scale 1-3

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

The description is specific, trigger-rich, and complete, clearly delineating the post-review merge-gate niche with an explicit Use-when / Not-for structure. It is a strong, concise skill description.

Dimension	Reasoning	Score
Specificity	Lists a concrete multi-step pipeline — "门禁检查 → PR → remote review → Feature Doc Truth 核对(pre-merge) → squash merge → 记录已合入(post-merge) → 清理" — naming multiple specific actions rather than vague language.	3 / 3
Completeness	Explicitly answers both what (the full merge pipeline) and when ("Use when: reviewer 放行后准备合入…"), and adds a "Not for" clause, satisfying the both-what-and-when requirement with explicit triggers.	3 / 3
Trigger Term Quality	Includes natural trigger terms users would say — "准备合入", "开 PR", "触发remote review", "准备 merge", plus the explicit triggers list ("合入 main", "merge", "gh pr create", "cloud review") — with good coverage of common phrasings.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear niche (post-approval merge flow) and the "Not for: 开发中、review 未通过、自检未完成" clause distinguishes it from quality-gate / request-review skills, making wrong-skill triggering unlikely.	3 / 3
	Total	12 / 12 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (663 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 9 missing	Warning

	Total	13 / 16 Passed

Repository: zts212653/clowder-ai
Path: cat-cafe-skills/merge-gate/SKILL.md
Commit: 80782c5

Reviewed: 11 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.