merge-gate

合入 main：按行为 / 数据 / 安全 / 契约 / 不可逆风险选择 targeted 或 full gate，并消费一个或多个有客观触发理由的独立 review source。

Quality

65%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./cat-cafe-skills/merge-gate/SKILL.md

Quality

Content

62%Weight 40%Scale 1-5

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

The body is an exceptionally well-sequenced, validation-rich merge workflow with highly actionable commands and feedback loops. Its main weakness is verbosity and a monolithic structure with no bundle files, leaving rationale prose and incident history inlined where tighter, split-out reference material would serve better.

Suggestions

Move incident histories and rationale prose (F168/F253 lessons, 病灶机制 explanations, 反面案例 PR #160) into a references/ file, keeping SKILL.md to the actionable procedure and rules.

Extract the Evidence Manifest field definitions, Provenance Matrix tables, and QA tables into a separate reference doc referenced one level deep, reducing the inline wall of text.

Tighten prose sections by replacing explanatory paragraphs with the rule + a one-line rationale, cutting repeated emphasis on the same constraint across sections.

Dimension	Reasoning	Score
Conciseness	The 739-line body is noticeably verbose, with extensive rationale prose ("为什么高风险需要这一步", "病灶机制"); incident citations (F168, F253, PR #160, PR #1300, operator 2026-07-15 投诉); and repeated emphasis that could be tightened without losing the actionable procedure.	2 / 5
Actionability	Core steps give concrete, copy-paste-ready bash (pnpm gate, gh pr create, gh pr comment --body '@codex review') and exact MCP call signatures with parameter tables; a few sections remain prose-only policy (Provenance Matrix judgment rules) that cannot be reduced to code.	4 / 5
Workflow Clarity	The process is explicitly sequenced (Lane 0, Steps 0.5→8.5) with hard validation checkpoints (Evidence Validation Checker E1-E5 BLOCKED conditions, fail-closed dirty-tree guards), feedback loops (CI repair state machine, 封板循环检测阈值), and checklists for complex parts.	5 / 5
Progressive Disclosure	No bundle files exist (references/scripts/assets absent), and the skill is a monolithic 739-line document; external references (../.cat-cafe-shared-refs/pr-template.md, pr-signals.md) are clearly signaled, but substantial inlined policy and QA detail that could live in separate reference files is not split out.	3 / 5
	Total	14 / 20 Passed

Description

67%Weight 40%Scale 1-5

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 concisely states what the skill does (risk-routed gate selection plus independent review source consumption) and when to use it (合入 main). It is specific and distinct, but trigger-term coverage within the description field is limited to one phrase and lacks an explicit Use-when clause.

Suggestions

Add an explicit '当...时使用 / Use when' clause enumerating concrete trigger phrases (合入 main, 准备 merge, 开 PR, gh pr create) directly in the description field.

Include common synonyms/file extensions or tool invocations (merge, PR, gh pr merge) so the description alone carries natural trigger terms rather than relying on the separate triggers list.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions — "选择 targeted 或 full gate" keyed to five named risk axes (行为/数据/安全/契约/不可逆) and "消费...独立 review source" — with only minor gaps in coverage of the full merge process.	4 / 5
Completeness	The "合入 main：" prefix supplies the when and the remainder supplies the what, so both are present; it stops short of an explicit "Use when..." clause with multiple concrete trigger phrases, so it is not a 5.	4 / 5
Trigger Term Quality	"合入 main" is a relevant natural trigger, but the description field itself omits common synonyms and variations (merge, PR, 开 PR) that users would say; those live in the separate triggers list rather than the description.	3 / 5
Distinctiveness Conflict Risk	The merge-gate niche (post-review合入流程, gate selection by risk) is clearly distinct from sibling skills like quality-gate and request-review, with only minor overlap risk on the generic "合入 main" trigger.	4 / 5
	Total	15 / 20 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 (739 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: 090626a

Reviewed: 1 day 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.