[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6WY04VACGDYDZ5AAJHTV1E1":3,"skill-same:01M6WY04VACGDYDZ5AAJHTV1E1":99},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":31,"publishedAt":52,"linkOnly":53,"markdownContent":54,"markdownBody":55,"frontMatter":56,"refName":78,"commitSha":79,"skillDirectory":80,"forksCount":14,"repositoryHtmlUrl":81,"repositoryHomepageUrl":82,"ownerHtmlUrl":83,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":84,"install":85,"resources":98},"01M6WY04VACGDYDZ5AAJHTV1E1","prd-writing",null,"Writes clear, actionable Product Requirements Documents — structured with problem statement, goals, user stories, scope, and success metrics. Use when asked to write a PRD, product spec, feature brief, or requirements document, or when designing a new feature that needs structured requirements.","projectious-work\u002Fprocesskit","projectious-work","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F263889726?v=4","01M6N97TJJ118WQ35CHRV8RD0E","projectious.work","01M6N97TPWJPF0BMKYQ4PBY5W8",0,"MIT","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Farchive\u002Fprocesskit-install-20261101\u002Fcontext\u002Ftemplates\u002Fprocesskit\u002Fv0.11.1\u002Fcontext\u002Fskills\u002Fproduct\u002Fprd-writing\u002FSKILL.md",[18,23,28],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000018","写作与文案","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000031","产品经理","PROFESSION","适用职业",{"categoryNo":29,"categoryName":30,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000036","内容创作者",[32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51],"writing","prd","prd writing","when","user stories","actionable","Product Requirements","problem","scope","project","artifact","wed","write","work","decision","open","brief","product","documents","workitem","2026-11-10 10:12:30",false,"---\nname: prd-writing\ndescription: |\n  Writes clear, actionable Product Requirements Documents — structured with problem statement, goals, user stories, scope, and success metrics. Use when asked to write a PRD, product spec, feature brief, or requirements document, or when designing a new feature that needs structured requirements.\nmetadata:\n  processkit:\n    apiVersion: processkit.projectious.work\u002Fv1\n    id: SKILL-prd-writing\n    version: \"1.1.0\"\n    created: 2026-04-08T00:00:00Z\n    category: product\n    layer: 2\n    uses:\n      - skill: workitem-management\n        purpose: Link requirements to WorkItems in the project backlog.\n      - skill: decision-record\n        purpose: Reference DecisionRecords that motivated requirements\n          or resolved open questions.\n    provides:\n      primitives: [Artifact]\n      mcp_tools: []\n      assets: []\n      processes: [prd-writing]\n---\n\n# PRD Writing\n\n## Intro\n\nA PRD (Product Requirements Document) aligns the team on what to\nbuild, why, and how to know when it's done. It is not a design\nspec, not a technical spec, and not a project plan — it is the\nbridge between user need and team execution. A good PRD can be read\nin 15 minutes and leave no major questions unanswered.\n\n## Overview\n\n### Feature PRD vs. product\u002Fvision PRD\n\nprocesskit distinguishes two PRD variants:\n\n**Feature PRD** — scopes a single feature or change. Audience: the\nimmediate delivery team. Contains: problem statement, use cases with\nacceptance criteria, functional + non-functional requirements,\nsuccess metrics. Template: see `### Feature PRD template` below.\n\n**Product\u002Fvision PRD** — defines the product itself: what it is, who\nit serves, what it must do, and how success is measured. Audience:\ncontributors, stakeholders, and agents picking up the project cold.\nContains: vision, target users, component map, core requirements,\nnon-goals, success metrics, milestones, glossary, open questions.\nTemplate: see `### Product PRD template` below.\n\n### Feature PRD template\n\n```markdown\n# [Feature Name] — Product Requirements\n\n**Status:** Draft | Review | Approved  \n**Owner:** [PM name]  \n**Target release:** [Version \u002F Quarter]  \n**Last updated:** [Date]\n\n---\n\n## Problem statement\n\n[2-3 sentences: what user problem is this solving? What evidence\nconfirms it's real? Why now?]\n\n## Goals\n\n**Primary goal:** [The single most important outcome — one sentence]\n\n**Secondary goals:**\n- [Additional measurable outcome]\n- [Additional measurable outcome]\n\n**Non-goals (explicitly out of scope):**\n- [What we are NOT building in this version]\n- [What we will address in a future iteration]\n\n## Users and use cases\n\n**Primary user:** [Role \u002F persona, with 1-sentence description]\n\n### Use cases\n\n**Use case 1: [Name]**\n> As a [user type], I want to [action] so that [benefit].\n\n- **Current experience:** [How users solve this today, or don't]\n- **Desired experience:** [What the new experience looks like]\n- **Acceptance criteria:**\n  - [ ] [Specific, testable criterion]\n  - [ ] [Specific, testable criterion]\n\n**Use case 2: [Name]**\n[Same structure]\n\n## Requirements\n\n### Functional requirements\n\n| ID | Requirement | Priority | Notes |\n|---|---|---|---|\n| REQ-001 | [Capability the product must have] | Must | |\n| REQ-002 | [Capability the product must have] | Should | |\n| REQ-003 | [Capability nice to have] | Could | |\n\nUse MoSCoW: Must \u002F Should \u002F Could \u002F Won't.\n\n### Non-functional requirements\n\n- **Performance:** [e.g., \"p99 response \u003C 500ms under 1000 concurrent users\"]\n- **Availability:** [e.g., \"99.9% uptime\"]\n- **Security:** [e.g., \"data encrypted at rest and in transit\"]\n- **Accessibility:** [e.g., \"WCAG 2.1 AA compliance\"]\n\n## Success metrics\n\n| Metric | Baseline | Target | Timeframe | How measured |\n|---|---|---|---|---|\n| [Primary metric] | [Current value] | [Goal] | [e.g., 90 days post-launch] | [Data source] |\n| [Secondary metric] | | | | |\n\n**Counter-metrics** (watch for regressions):\n- [Metric that should not worsen]\n\n## Dependencies and constraints\n\n- **Technical dependencies:** [APIs, infrastructure, other teams' work]\n- **Design dependency:** [Design work required before development]\n- **Timeline constraint:** [Hard deadlines and their reason]\n- **Regulatory constraint:** [Compliance requirements]\n\n## Open questions\n\n| Question | Owner | Due by | Status |\n|---|---|---|---|\n| [Decision needed] | [Name] | [Date] | Open |\n\n## Appendix\n\n[Links to: user research, design mockups, competitive analysis,\nanalytics data, technical feasibility notes]\n```\n\n### Product PRD template\n\nFor product-level or vision-level PRDs:\n\n~~~markdown\n# [Product Name] — Product Requirements\n\n**Status:** Draft | Review | Approved\n**Owner:** [name or team]\n**Last updated:** [YYYY-MM-DD]\n\n---\n\n## Vision\n\n[2-4 sentences: what is this product, who is it for, and why does\nit exist? Distinct from goals — this is the directional statement.]\n\n## Problem statement\n\n[What problem does this product solve? What evidence confirms it is\nreal? Why is this the right solution?]\n\n## Goals\n\n**Primary goal:** [The single most important outcome.]\n\n**Secondary goals:**\n- [Measurable secondary outcome]\n\n**Non-goals:** See Non-goals section.\n\n## Target users\n\n| User | What they get |\n|---|---|\n| [Role] | [Value] |\n\n## Component map\n\n[ASCII diagram or description of how major components relate.]\n\n## Core requirements\n\n### R1 — [Requirement name]\n\n[Requirement description. Reference the format spec or schema where\napplicable rather than reproducing it inline.]\n\n### R2 — ...\n\n## Non-functional requirements\n\n- **Portability:** ...\n- **Offline capability:** ...\n- **Backward compatibility:** ...\n\n## Non-goals\n\n- [Explicitly out of scope]\n\n## Success metrics\n\n| Metric | Status | Target |\n|---|---|---|\n| [Metric] | [Current] | [Goal] |\n\n## Milestones\n\n[What has shipped. What v1.0 \u002F next milestone requires. No date\ncommitments unless hard deadlines exist.]\n\n## Glossary\n\n| Term | Definition |\n|---|---|\n| [Term] | [One-line definition] |\n\n## Open questions and constraints\n\n**Technical constraints:**\n- [Hard constraint]\n\n**Open design decisions:**\n- [Decision not yet made]\n\n## Appendix\n\n- [Link to canonical spec files, related decisions, research]\n~~~\n\n### Before writing — gather the inputs\n\nA PRD written without evidence is advocacy, not requirements. Gather\nfirst:\n\n1. **User pain**: interviews, support tickets, NPS feedback, usage\n   data showing the gap\n2. **Business case**: why this over other options? Revenue impact,\n   churn risk, strategic fit\n3. **Technical feasibility**: rough signal from engineering that\n   the approach is buildable in the target timeframe\n4. **Competitive context**: how do alternatives solve this?\n\n### Scoping: non-goals are as important as goals\n\nEvery PRD should explicitly list what is NOT in scope. Non-goals:\n- Prevent scope creep during development\n- Let the team build faster (smaller, clearer target)\n- Force the PM to actually choose priorities\n\nIf every obvious extension is in-scope, the PRD is not prioritized.\n\n### Writing acceptance criteria\n\nEach use case must have testable acceptance criteria — not \"works\nwell\" but \"the user can complete [task] without [error]\".\n\n**Bad:** \"The export works correctly.\"  \n**Good:**\n- [ ] User can export up to 10,000 rows as CSV in \u003C 5 seconds\n- [ ] Export includes all columns visible on the current filter\n- [ ] Filename defaults to `[report-name]-[YYYY-MM-DD].csv`\n\n### Keeping the PRD current\n\nA PRD that is written once and never updated becomes a liability.\nMark decisions made, questions answered, and requirements changed.\nA \"Status: Approved\" PRD with open questions that were resolved\nwithout updating the document causes downstream confusion.\n\n## Gotchas\n\nAgent-specific failure modes — provider-neutral pause-and-self-check items:\n\n- **Writing solution first, problem second.** A PRD that opens with \"We will build a dashboard\" skips the why — it's a solution looking for a problem. Always lead with the user problem and the evidence that it's real. The solution follows from the problem, not the other way around.\n- **Writing acceptance criteria that are not testable.** \"The UI is intuitive\" is not an acceptance criterion — it cannot be verified. \"A new user can complete the primary task in under 2 minutes without asking for help\" is testable. Each acceptance criterion must have a binary pass\u002Ffail.\n- **Omitting non-goals.** Without explicit non-goals, the team will build the obvious extensions and the feature will take 3× as long. Non-goals force a prioritization decision and give the team permission to say \"that's out of scope\" during development.\n- **Writing requirements that prescribe implementation instead of behavior.** \"Use a dropdown\" is an implementation decision; \"the user selects a single option from the list\" is a requirement. PRDs should describe what the product does, not how it's built — that's the team's decision.\n- **Skipping success metrics.** A PRD without success metrics produces a feature you can ship but never evaluate. If you cannot define what \"success\" looks like before building, you cannot tell after shipping whether the investment was worth it.\n- **Leaving open questions open at launch.** Questions that are open at PRD approval must be answered before development starts. A PRD with open architecture questions or unresolved design decisions is not ready for development — it will be decided ad-hoc during implementation, usually badly.\n- **Making the PRD too long.** A PRD that takes 2 hours to read will not be read by the team before development starts. Keep it to one focused session (15-30 min) — the full detail belongs in linked design docs, technical specs, and research reports, not in the PRD itself.\n- **Writing a PRD without loading this skill first.** The trigger\n  phrase \"PRD\" or \"product requirements\" should load this skill\n  before any writing begins. Writing from general knowledge produces\n  a document that may look correct but misses processkit-specific\n  conventions (Artifact storage, workitem linking, feature vs.\n  product PRD distinction).\n\n## Full reference\n\n### PRD vs. technical spec vs. design spec\n\n| Document | Owner | Answers | When written |\n|---|---|---|---|\n| PRD | Product manager | What and why | Before design and eng start |\n| Design spec | Designer | How it looks and feels | After PRD approval |\n| Technical spec | Engineering | How it's built | After design spec |\n\nThe PRD is the source of truth for requirements. Design and technical\nspecs are downstream — they should be consistent with the PRD, and\ncontradictions should trigger a PRD update, not a silent departure.\n\n### MoSCoW prioritization\n\n| Priority | Meaning | Implication |\n|---|---|---|\n| **Must** | The feature cannot ship without this | Block release if missing |\n| **Should** | High value but can be worked around temporarily | Include if time allows; don't block |\n| **Could** | Nice to have, low risk if cut | Cuts first when scope must shrink |\n| **Won't** | Explicitly out of scope for this version | Documents a conscious choice |\n\n### Linking to processkit context\n\nWhen the project uses processkit, link the PRD to:\n- The WorkItem(s) it spawns (list BACK-NNN IDs in the appendix)\n- Relevant decisions (`context\u002Fdecisions\u002F` or DECISIONS.md)\n- Research notes (Note entities with `type: reference`)\n\nSave the PRD as `context\u002Fartifacts\u002Fprd-[feature-name]-[date].md` or\nlink it as an Artifact entity in the index.\n\nThe PRD itself should be stored as an `Artifact` entity under\n`context\u002Fartifacts\u002Fprd-\u003Cproduct-or-feature>-\u003Cdate>.md` so it is\nindexed and queryable. Create the entity with:\n\n    spec:\n      name: \u003Cdescriptive name>\n      kind: document\n      location: context\u002Fartifacts\u002Fprd-\u003Cname>-\u003Cdate>.md\n      format: markdown\n      tags: [prd]\n","# PRD Writing\n\n## Intro\n\nA PRD (Product Requirements Document) aligns the team on what to\nbuild, why, and how to know when it's done. It is not a design\nspec, not a technical spec, and not a project plan — it is the\nbridge between user need and team execution. A good PRD can be read\nin 15 minutes and leave no major questions unanswered.\n\n## Overview\n\n### Feature PRD vs. product\u002Fvision PRD\n\nprocesskit distinguishes two PRD variants:\n\n**Feature PRD** — scopes a single feature or change. Audience: the\nimmediate delivery team. Contains: problem statement, use cases with\nacceptance criteria, functional + non-functional requirements,\nsuccess metrics. Template: see `### Feature PRD template` below.\n\n**Product\u002Fvision PRD** — defines the product itself: what it is, who\nit serves, what it must do, and how success is measured. Audience:\ncontributors, stakeholders, and agents picking up the project cold.\nContains: vision, target users, component map, core requirements,\nnon-goals, success metrics, milestones, glossary, open questions.\nTemplate: see `### Product PRD template` below.\n\n### Feature PRD template\n\n```markdown\n# [Feature Name] — Product Requirements\n\n**Status:** Draft | Review | Approved  \n**Owner:** [PM name]  \n**Target release:** [Version \u002F Quarter]  \n**Last updated:** [Date]\n\n---\n\n## Problem statement\n\n[2-3 sentences: what user problem is this solving? What evidence\nconfirms it's real? Why now?]\n\n## Goals\n\n**Primary goal:** [The single most important outcome — one sentence]\n\n**Secondary goals:**\n- [Additional measurable outcome]\n- [Additional measurable outcome]\n\n**Non-goals (explicitly out of scope):**\n- [What we are NOT building in this version]\n- [What we will address in a future iteration]\n\n## Users and use cases\n\n**Primary user:** [Role \u002F persona, with 1-sentence description]\n\n### Use cases\n\n**Use case 1: [Name]**\n> As a [user type], I want to [action] so that [benefit].\n\n- **Current experience:** [How users solve this today, or don't]\n- **Desired experience:** [What the new experience looks like]\n- **Acceptance criteria:**\n  - [ ] [Specific, testable criterion]\n  - [ ] [Specific, testable criterion]\n\n**Use case 2: [Name]**\n[Same structure]\n\n## Requirements\n\n### Functional requirements\n\n| ID | Requirement | Priority | Notes |\n|---|---|---|---|\n| REQ-001 | [Capability the product must have] | Must | |\n| REQ-002 | [Capability the product must have] | Should | |\n| REQ-003 | [Capability nice to have] | Could | |\n\nUse MoSCoW: Must \u002F Should \u002F Could \u002F Won't.\n\n### Non-functional requirements\n\n- **Performance:** [e.g., \"p99 response \u003C 500ms under 1000 concurrent users\"]\n- **Availability:** [e.g., \"99.9% uptime\"]\n- **Security:** [e.g., \"data encrypted at rest and in transit\"]\n- **Accessibility:** [e.g., \"WCAG 2.1 AA compliance\"]\n\n## Success metrics\n\n| Metric | Baseline | Target | Timeframe | How measured |\n|---|---|---|---|---|\n| [Primary metric] | [Current value] | [Goal] | [e.g., 90 days post-launch] | [Data source] |\n| [Secondary metric] | | | | |\n\n**Counter-metrics** (watch for regressions):\n- [Metric that should not worsen]\n\n## Dependencies and constraints\n\n- **Technical dependencies:** [APIs, infrastructure, other teams' work]\n- **Design dependency:** [Design work required before development]\n- **Timeline constraint:** [Hard deadlines and their reason]\n- **Regulatory constraint:** [Compliance requirements]\n\n## Open questions\n\n| Question | Owner | Due by | Status |\n|---|---|---|---|\n| [Decision needed] | [Name] | [Date] | Open |\n\n## Appendix\n\n[Links to: user research, design mockups, competitive analysis,\nanalytics data, technical feasibility notes]\n```\n\n### Product PRD template\n\nFor product-level or vision-level PRDs:\n\n~~~markdown\n# [Product Name] — Product Requirements\n\n**Status:** Draft | Review | Approved\n**Owner:** [name or team]\n**Last updated:** [YYYY-MM-DD]\n\n---\n\n## Vision\n\n[2-4 sentences: what is this product, who is it for, and why does\nit exist? Distinct from goals — this is the directional statement.]\n\n## Problem statement\n\n[What problem does this product solve? What evidence confirms it is\nreal? Why is this the right solution?]\n\n## Goals\n\n**Primary goal:** [The single most important outcome.]\n\n**Secondary goals:**\n- [Measurable secondary outcome]\n\n**Non-goals:** See Non-goals section.\n\n## Target users\n\n| User | What they get |\n|---|---|\n| [Role] | [Value] |\n\n## Component map\n\n[ASCII diagram or description of how major components relate.]\n\n## Core requirements\n\n### R1 — [Requirement name]\n\n[Requirement description. Reference the format spec or schema where\napplicable rather than reproducing it inline.]\n\n### R2 — ...\n\n## Non-functional requirements\n\n- **Portability:** ...\n- **Offline capability:** ...\n- **Backward compatibility:** ...\n\n## Non-goals\n\n- [Explicitly out of scope]\n\n## Success metrics\n\n| Metric | Status | Target |\n|---|---|---|\n| [Metric] | [Current] | [Goal] |\n\n## Milestones\n\n[What has shipped. What v1.0 \u002F next milestone requires. No date\ncommitments unless hard deadlines exist.]\n\n## Glossary\n\n| Term | Definition |\n|---|---|\n| [Term] | [One-line definition] |\n\n## Open questions and constraints\n\n**Technical constraints:**\n- [Hard constraint]\n\n**Open design decisions:**\n- [Decision not yet made]\n\n## Appendix\n\n- [Link to canonical spec files, related decisions, research]\n~~~\n\n### Before writing — gather the inputs\n\nA PRD written without evidence is advocacy, not requirements. Gather\nfirst:\n\n1. **User pain**: interviews, support tickets, NPS feedback, usage\n   data showing the gap\n2. **Business case**: why this over other options? Revenue impact,\n   churn risk, strategic fit\n3. **Technical feasibility**: rough signal from engineering that\n   the approach is buildable in the target timeframe\n4. **Competitive context**: how do alternatives solve this?\n\n### Scoping: non-goals are as important as goals\n\nEvery PRD should explicitly list what is NOT in scope. Non-goals:\n- Prevent scope creep during development\n- Let the team build faster (smaller, clearer target)\n- Force the PM to actually choose priorities\n\nIf every obvious extension is in-scope, the PRD is not prioritized.\n\n### Writing acceptance criteria\n\nEach use case must have testable acceptance criteria — not \"works\nwell\" but \"the user can complete [task] without [error]\".\n\n**Bad:** \"The export works correctly.\"  \n**Good:**\n- [ ] User can export up to 10,000 rows as CSV in \u003C 5 seconds\n- [ ] Export includes all columns visible on the current filter\n- [ ] Filename defaults to `[report-name]-[YYYY-MM-DD].csv`\n\n### Keeping the PRD current\n\nA PRD that is written once and never updated becomes a liability.\nMark decisions made, questions answered, and requirements changed.\nA \"Status: Approved\" PRD with open questions that were resolved\nwithout updating the document causes downstream confusion.\n\n## Gotchas\n\nAgent-specific failure modes — provider-neutral pause-and-self-check items:\n\n- **Writing solution first, problem second.** A PRD that opens with \"We will build a dashboard\" skips the why — it's a solution looking for a problem. Always lead with the user problem and the evidence that it's real. The solution follows from the problem, not the other way around.\n- **Writing acceptance criteria that are not testable.** \"The UI is intuitive\" is not an acceptance criterion — it cannot be verified. \"A new user can complete the primary task in under 2 minutes without asking for help\" is testable. Each acceptance criterion must have a binary pass\u002Ffail.\n- **Omitting non-goals.** Without explicit non-goals, the team will build the obvious extensions and the feature will take 3× as long. Non-goals force a prioritization decision and give the team permission to say \"that's out of scope\" during development.\n- **Writing requirements that prescribe implementation instead of behavior.** \"Use a dropdown\" is an implementation decision; \"the user selects a single option from the list\" is a requirement. PRDs should describe what the product does, not how it's built — that's the team's decision.\n- **Skipping success metrics.** A PRD without success metrics produces a feature you can ship but never evaluate. If you cannot define what \"success\" looks like before building, you cannot tell after shipping whether the investment was worth it.\n- **Leaving open questions open at launch.** Questions that are open at PRD approval must be answered before development starts. A PRD with open architecture questions or unresolved design decisions is not ready for development — it will be decided ad-hoc during implementation, usually badly.\n- **Making the PRD too long.** A PRD that takes 2 hours to read will not be read by the team before development starts. Keep it to one focused session (15-30 min) — the full detail belongs in linked design docs, technical specs, and research reports, not in the PRD itself.\n- **Writing a PRD without loading this skill first.** The trigger\n  phrase \"PRD\" or \"product requirements\" should load this skill\n  before any writing begins. Writing from general knowledge produces\n  a document that may look correct but misses processkit-specific\n  conventions (Artifact storage, workitem linking, feature vs.\n  product PRD distinction).\n\n## Full reference\n\n### PRD vs. technical spec vs. design spec\n\n| Document | Owner | Answers | When written |\n|---|---|---|---|\n| PRD | Product manager | What and why | Before design and eng start |\n| Design spec | Designer | How it looks and feels | After PRD approval |\n| Technical spec | Engineering | How it's built | After design spec |\n\nThe PRD is the source of truth for requirements. Design and technical\nspecs are downstream — they should be consistent with the PRD, and\ncontradictions should trigger a PRD update, not a silent departure.\n\n### MoSCoW prioritization\n\n| Priority | Meaning | Implication |\n|---|---|---|\n| **Must** | The feature cannot ship without this | Block release if missing |\n| **Should** | High value but can be worked around temporarily | Include if time allows; don't block |\n| **Could** | Nice to have, low risk if cut | Cuts first when scope must shrink |\n| **Won't** | Explicitly out of scope for this version | Documents a conscious choice |\n\n### Linking to processkit context\n\nWhen the project uses processkit, link the PRD to:\n- The WorkItem(s) it spawns (list BACK-NNN IDs in the appendix)\n- Relevant decisions (`context\u002Fdecisions\u002F` or DECISIONS.md)\n- Research notes (Note entities with `type: reference`)\n\nSave the PRD as `context\u002Fartifacts\u002Fprd-[feature-name]-[date].md` or\nlink it as an Artifact entity in the index.\n\nThe PRD itself should be stored as an `Artifact` entity under\n`context\u002Fartifacts\u002Fprd-\u003Cproduct-or-feature>-\u003Cdate>.md` so it is\nindexed and queryable. Create the entity with:\n\n    spec:\n      name: \u003Cdescriptive name>\n      kind: document\n      location: context\u002Fartifacts\u002Fprd-\u003Cname>-\u003Cdate>.md\n      format: markdown\n      tags: [prd]\n",{"name":5,"metadata":57,"description":77},{"processkit":58},{"id":59,"uses":60,"layer":67,"created":68,"version":69,"category":49,"provides":70,"apiVersion":76},"SKILL-prd-writing",[61,64],{"skill":62,"purpose":63},"workitem-management","Link requirements to WorkItems in the project backlog.",{"skill":65,"purpose":66},"decision-record","Reference DecisionRecords that motivated requirements or resolved open questions.",2,1775606400000,"1.1.0",{"assets":71,"mcp_tools":72,"processes":73,"primitives":74},[],[],[5],[75],"Artifact","processkit.projectious.work\u002Fv1","Writes clear, actionable Product Requirements Documents — structured with problem statement, goals, user stories, scope, and success metrics. Use when asked to write a PRD, product spec, feature brief, or requirements document, or when designing a new feature that needs structured requirements.\n","main","4876a92620ee792aac2ac9a1a806322d8a572367","archive\u002Fprocesskit-install-20261101\u002Fcontext\u002Ftemplates\u002Fprocesskit\u002Fv0.11.1\u002Fcontext\u002Fskills\u002Fproduct\u002Fprd-writing","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit","https:\u002F\u002Fprojectious-work.github.io\u002Fprocesskit\u002F","https:\u002F\u002Fgithub.com\u002Fprojectious-work","2026-11-11 23:24:18",{"source":8,"skillName":5,"commands":86,"prompt":96,"zipAvailable":97},[87,90,93],{"tool":88,"command":89},"npx","npx skills add projectious-work\u002Fprocesskit --skill prd-writing",{"tool":91,"command":92},"bunx","bunx skills add projectious-work\u002Fprocesskit --skill prd-writing",{"tool":94,"command":95},"pnpm","pnpm dlx skills add projectious-work\u002Fprocesskit --skill prd-writing","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Farchive\u002Fprocesskit-install-20261101\u002Fcontext\u002Ftemplates\u002Fprocesskit\u002Fv0.11.1\u002Fcontext\u002Fskills\u002Fproduct\u002Fprd-writing\u002FSKILL.md",true,[],[100,107,114,120,145,164],{"skillNo":101,"name":102,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":105,"publishedAt":106},"01M7665G2ED3624R3X93304265","standard-ui-design","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fstandard-ui-design\u002FSKILL.md",[],[],"2026-11-12 13:55:43",{"skillNo":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":111,"categories":112,"keywords":113,"publishedAt":106},"01M76651QTC6W0HJ4AFW2MMR3V","standard-agent-use","Company standard for working with AI agents. Use when deciding what to delegate to agents, capturing open work, keeping agent memory or knowledge notes, improving an agent workflow, or introducing agent work in a team or for a client.","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fstandard-agent-use\u002FSKILL.md",[],[],{"skillNo":115,"name":116,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":117,"categories":118,"keywords":119,"publishedAt":106},"01M76650NN4T8P241021NBR64S","standard-agent-building","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fstandard-agent-building\u002FSKILL.md",[],[],{"skillNo":121,"name":122,"nameZh":6,"description":123,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":124,"categories":125,"keywords":129,"publishedAt":52},"01M6X5MTC89B4ZGXW3JVSQCPF8","skill-management","Manage Skill package manifests as v1 ontology entities.","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fprocesskit\u002Fskill-management\u002FSKILL.md",[126],{"categoryNo":127,"categoryName":128,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[130,131,132,133,134,135,45,136,137,138,139,140,122,141,142,143,144],"management","skill management","ontology","v1","create","as","transition","get","manage","update","alpha","list","package","skill package","manifests",{"skillNo":146,"name":147,"nameZh":6,"description":148,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":149,"categories":150,"keywords":152,"publishedAt":52},"01M6X5MCV74NHW5MEDNHZRF2C8","schema-management","Manage generated processkit schema contracts through deterministic MCP rebuild and inspection workflows.","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fprocesskit\u002Fschema-management\u002FSKILL.md",[151],{"categoryNo":127,"categoryName":128,"taxonomyType":21,"taxonomyTypeDesc":22},[153,154,130,155,156,157,158,43,159,160,161,162,45,137,138,140,163],"schema","schema management","contract","rebuild","contracts","workflows","generated","MCP","validation","deterministic","mode",{"skillNo":165,"name":166,"nameZh":6,"description":167,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":168,"categories":169,"keywords":170,"publishedAt":52},"01M6X5M5G6YJ96AMZGSCZJ1D13","proposition-management","Record claims, risks, beliefs, world facts, WSJF estimates, and assumptions through the Proposition interface.","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fprocesskit\u002Fproposition-management\u002FSKILL.md",[],[171,130,172,173,134,174,175,176,45,137,177,178,139,179,140,180,181,182,183,184],"proposition management","proposition","interface","claims","record","world","creation","event","WSJF","index","updates","query","facts","log"]