write-mrd

输入一个产品 idea（一句话或详细描述均可），自动生成图文并茂、结构完整的 MRD（市场需求文档），以 Markdown 格式保存到当前项目 markdown/ 目录，同时输出配套 SVG 图表文件。触发条件：用户提到"写 MRD"、"帮我写市场需求文档"、"生成 MRD"、"写产品市场需求"、"我有个产品想法"、"帮我把这个 idea 写成 MRD"、"市场需求文档模板"、"产品立项文档"、"帮我写需求文档"。即使用户只给出一句简短的产品描述如"我想做一个 XX 产品"，也应立即使用本 skill 生成 MRD。本 skill 面向产品经理、创业者、产品负责人，输出可直接用于团队评审和立项决策的专业文档。

Quality

80%

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

Quality

Content

61%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 skill body is well-structured and highly actionable with concrete templates, naming conventions, and a self-check checklist. Its main weaknesses are length/over-explanation of concepts Claude already knows, a missing explicit validation feedback loop for batch file-writing operations (capping workflow clarity at 3), and total reliance on a single inlined file with no progressive disclosure into bundle references.

Suggestions

Add an explicit validation checkpoint in the output workflow (e.g. after Step 4/5: verify SVG files render and are correctly referenced in the Markdown, fix and regenerate if not) to lift workflow clarity above 3.

Tighten or remove explanatory prose Claude already knows (JTBD three-task definitions, RICE formula, North Star metric concept) to improve conciseness; keep the templates and checklists which earn their tokens.

Split the bulky template specs (3.x sections) and the SVG style guide (4.3) into reference files under references/ and link to them one level deep, so SKILL.md becomes a lean overview.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient with templates and code blocks rather than prose, but at ~320 lines it is lengthy and includes some over-explanation Claude does not need (e.g. explaining JTBD three layers, RICE formula, '北极星指标' definitions). It could be tightened without losing actionability.	3 / 5
Actionability	Provides concrete, copy-paste-ready templates (document header block, user-role card, milestone table, SVG viewBox/font/color spec, file naming) and a self-check checklist. Minor gaps: web-search guidance uses query-string hints rather than exact commands, and SVG generation is specified by spec rather than executable code.	4 / 5
Workflow Clarity	A clear 5-step sequence is present with a self-check checklist at the end, but validation checkpoints within the flow are mostly implicit. Per the rubric's batch/destructive cap, file-writing and SVG-generation batch operations lack an explicit validate->fix->retry feedback loop, capping this dimension at 3.	3 / 5
Progressive Disclosure	No bundle files exist (references/, scripts/, assets/ absent), so all content is inlined in SKILL.md. The document is well-sectioned internally, but content that could live in separate files (full template specs, SVG style guide) is inlined with no one-level-deep references, placing it at the 'some structure, could be better organized' anchor.	3 / 5
	Total	13 / 20 Passed

Description

100%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 is comprehensive, specifying concrete actions (MRD generation, Markdown output, SVG charts), an explicit trigger-conditions clause with many natural phrasings, and clear target audience. It fully answers both 'what' and 'when'. Minor weakness is mild verbosity and the slightly broad '我有个产品想法' trigger, but distinctiveness is strong.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions with comprehensive coverage: '生成图文并茂、结构完整的 MRD', '以 Markdown 格式保存到当前项目 markdown/ 目录，同时输出配套 SVG 图表文件', naming the domain (MRD generation), the output format (Markdown), output location (markdown/ directory), and SVG chart generation.	5 / 5
Completeness	Clearly and explicitly answers both 'what' (generate structured illustrated MRD in Markdown with SVG charts to markdown/ directory) and 'when' (explicit '触发条件：...' clause with concrete trigger phrases and the short-phrase rule), with concrete trigger phrases present.	5 / 5
Trigger Term Quality	Comprehensive coverage of natural trigger phrases users would say including synonyms and exact phrasings: '写 MRD', '帮我写市场需求文档', '生成 MRD', '写产品市场需求', '我有个产品想法', '帮我把这个 idea 写成 MRD', '市场需求文档模板', '产品立项文档', '帮我写需求文档', plus the short-phrase case '我想做一个 XX 产品'.	5 / 5
Distinctiveness Conflict Risk	Clear niche (MRD/市场需求文档 generation) with distinct triggers; the combination of MRD, product idea, and 立项文档 triggers is specific and unlikely to conflict with other skills. Minor caveat: '我有个产品想法' is somewhat broad but the surrounding context pins it to MRD generation.	5 / 5
	Total	20 / 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
relative_links	Relative link issues: 1 missing	Warning

	Total	15 / 16 Passed

Repository: digoal/blog
Path: skills/skills_for_claude_web/write-mrd/SKILL.md
Commit: 3b9c83d

Reviewed: 3 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.