writing-plans

把已接受的愿景与需求转成接手者可实施、可验收的计划。 Use when: 跨组件、状态对象、依赖或实施顺序需要先理清，且没有足够的现成计划。 Not for: 路径已明确的简单改动、已有可执行计划、尚未接受的愿景决策。 Output: 目标与范围、关键契约/依赖、可执行工作单元和验证方式。

Quality

83%

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

82%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 content is an efficient, highly actionable instruction skill: concrete commands, a usable plan template, an explicit state-contract gate, and a mistake table replace abstraction with executable guidance. Remaining gaps are minor — some redundancy between sections and a leftover removed-reference placeholder.

Suggestions

Remove or resolve the dangling "*(internal reference removed)*" placeholder in the Stateful Object Gate section, since it currently points nowhere.

Deduplicate the 常见误区 table against the 必须做到 bullets (e.g., step granularity appears in both) to tighten the token budget.

Consider making the per-work-unit validation loop explicit (verify → fix → re-verify) in the optional 4-step method, rather than only describing validation as a requirement.

Dimension	Reasoning	Score
Conciseness	The body is dense and assumes Claude's competence with no beginner-concept padding, but the 常见误区 table partially repeats the 必须做到 bullets (e.g., step-granularity guidance appears in both) and the 可选参考 section overlaps the must-do list, so minor trimming is possible.	4 / 5
Actionability	It provides a copy-paste markdown plan skeleton, concrete verified commands ("pnpm check:fix", "pnpm biome format --write <files>", "pnpm check"), a concrete 4-step decomposition method, and a mistake-vs-correct-practice table — fully usable guidance covering the common cases.	5 / 5
Workflow Clarity	The 从终态拆工作 section gives a clear 4-step sequence and validation is woven throughout (per-work-unit observable outcomes, the 验证 field in the skeleton, terminal verification via "pnpm check"), but checkpoints are stated as requirements rather than an explicit ordered validate-and-recover flow.	4 / 5
Progressive Disclosure	At ~75 lines with well-organized sections, a table, and a template, the body needs no external references and navigates well; however, the dangling "(internal reference removed)" placeholder is a small blemish and a couple of sections (e.g., Formatting Command Contract) sit at the edge of extractable size.	4 / 5
	Total	17 / 20 Passed

Description

83%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.

A strong description: it explicitly states what it does, when to use it, when not to use it, and what the output looks like, with concrete scenario-based triggers. The only weaknesses are modest synonym coverage for trigger terms and slight overlap risk with other planning skills.

Dimension	Reasoning	Score
Specificity	The description names one concrete core action ("转成接手者可实施、可验收的计划") and enumerates specific deliverables ("目标与范围、关键契约/依赖、可执行工作单元和验证方式"), which is more than 1-2 actions but stops short of a list of multiple distinct actions.	4 / 5
Completeness	It explicitly answers both what ("把已接受的愿景与需求转成...计划") and when ("Use when: ...且没有足够的现成计划"), further sharpened by an explicit "Not for" boundary and an "Output" clause.	5 / 5
Trigger Term Quality	"跨组件、状态对象、依赖或实施顺序需要先理清" plus plan-related terms give good coverage of natural triggering scenarios users would express, though common synonyms and variations are missing.	4 / 5
Distinctiveness Conflict Risk	The "Not for" clause (simple clear-path changes, existing executable plans, unaccepted vision decisions) carves out a clear niche, but generic trigger phrases like "implementation plan" carry minor overlap risk with other planning-related skills.	4 / 5
	Total	17 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: zts212653/clowder-ai
Path: cat-cafe-skills/writing-plans/SKILL.md
Commit: 61389cc

Reviewed: 2 hours 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.