prd-writing
Skill4 starsChanged 3 months agoMITView on GitHub
Run a structured discovery interview and produce a complete, developer-ready Product Requirements Document. Load when the user asks to write a PRD, create product requirements, document a feature, define user stories with acceptance criteria, or turn a rough idea into a formal product requirements document. Also triggers on "document this feature", "write requirements for", "create a one-pager", "turn this into a PRD", "I need a PRD for", or any request to produce a structured product document for stakeholder alignment or engineering handoff. Supports Full PRD, Lean PRD, and One-Pager formats. Note: for executable feature specifications (FRs, NFRs, ACs as Given/When/Then consumable by AI coding agents), route to `feature-spec` instead — PRDs frame the product, feature-specs encode the implementable contract.
17 other projects use a version of this.
Adopted by
Most recently taken up by
- okou-ai/okou-skillstheir own version, first seen Nov 11, 2026
- SWEStash/swe-workflow-skillstheir own version, first seen Nov 8, 2026
- jokob-sk/NetAlertXtheir own version, first seen Nov 1, 2026
- netalertx/NetAlertXtheir own version, first seen Nov 1, 2026
- Maudeunfledged834/startup-founder-skillstheir own version, first seen Oct 25, 2026
- davidanthonyn/claude-skillstheir own version, first seen Oct 20, 2026
Used with
Projects with it also use feedback-synthesis, onboarding-flow and 2 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
Dragoon0xSkillfeedback-synthesis
Turn raw customer feedback from multiple sources into clear product insights and priorities. Covers feedback collection, pattern recognition, and separating signal from noise. Use when feedback feels overwhelming, contradictory, or when the team can't agree on what users actually want.
Also in 3 projects that use this
Dragoon0xSkillonboarding-flow
Design first-run experiences that create the aha moment fast. Reduces time-to-value by sequencing actions, progressive disclosure, and contextual guidance. Use when building signup flows, product tours, or empty states.
Also in 3 projects that use this
Microsoft AzureSkillgit-workflow
Squad branching model: dev-first workflow with insiders preview channel
Also in 3 projects that use this
- +1 more from Dragoon0x
Other versions17 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 17
- jokob-sk/NetAlertX★ 7.4ksame name
- netalertx/NetAlertX★ 7.4ksame name
- shawnpang/startup-founder-skills★ 331same name
- GenerativeAgents/claude-code-book-chapter8★ 114same name
- itseffi/agentic-os★ 114same name
- okou-ai/okou-skills★ 81same name
- vm0-ai/vm0-skills★ 81same name
- Maudeunfledged834/startup-founder-skills★ 8same name
- Dragoon0x/product-skills★ 4same name
- mattstruble/skills★ 3same name
7 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository125 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .cursor/rules/agent-loom-routing.mdcCursor rule
- .cursor/rules/agent-loom-skills-index.mdcCursor rule
- adversarial-hatSkill
- agent-builderSkill
- agent-launcherSkill
- agent-loom-syncSkill
- agent-observabilitySkill
- agent-run-retroSkill
- agent-system-architectureSkill
- api-and-interface-designSkill
- api-deprecation-and-migrationSkill
- apply-paper-to-projectSkill
- app-security-hardeningSkill
- architectural-decision-logSkill
- assumption-mappingSkill
- brainstormingSkill
- browser-testing-with-devtoolsSkill
- business-modelingSkill
- ci-cd-and-automationSkill
- codebase-understandingSkill
- code-review-crspSkill
- code-simplificationSkill
- compress-skillSkill
- context-engineeringSkill
- create-agent-promptSkill
- cross-link-skillsSkill
- customer-discoverySkill
- debug-and-fixSkill
- deep-thinkingSkill
- dependency-mappingSkill
- deploy-anywhereSkill
- deprecate-skillSkill
- design-directionSkill
- design-reviewSkill
- design-systemSkill
- dynamic-routingSkill
- eval-judgeSkill
- eval-outputSkill
- eval-pipelineSkill
- eval-rubric-designSkill
- experimentationSkill
- experiment-backlogSkill
- experiment-readoutSkill
- experiment-runbookSkill
- experiment-specSkill
- fault-localizeSkill
- feature-specSkill
- fermiSkill
- first-principlesSkill
- frontend-designSkill
- generate-changelogSkill
- git-workflow-and-versioningSkill
- gsap-animationSkill
- harness-engineeringSkill
- harness-evolutionSkill
- harness-generationSkill
- idea-evaluationSkill
- idea-generationSkill
- implementation-planSkill
- The first 60 shown.
The file
Show its rules and text
We split this skill into the rules it gives an agent
17 rules. 0 are common, 9 are distinctive, and 8 are very specific to this project.
0
Common
Most of the 13 edited versions we read keep these.
9
Distinctive
This author's own choices. Worth a look.
8
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 34
Never write the PRD before asking at least 2 discovery questions — even with a detailed brief.
- Very specific: line 35
Never use "fast", "easy", "intuitive", "modern", "robust" — replace with specific, testable criteria or mark TBD — confirm with engineering.
- Very specific: line 36
Never mark a PRD Approved if the Open Questions section is non-empty.
- Very specific: line 37
Never hallucinate a tech stack — if unspecified, write TBD.
- Distinctive: line 44
Look for, in priority order:
- Distinctive: line 48
Ask only about what's missing after reading available context.
- Distinctive: line 51
Ask or infer: Full PRD · Lean PRD · One-Pager · Technical PRD.
- Distinctive: line 58
Must answer before writing:
- Distinctive: line 66
Before writing, offer: "Shall I run deep-thinking on the requirements?"
- Very specific: line 75
Use the schema for the chosen format from references/prd-schemas.md.
- Distinctive: line 97
Ask: "Which sections need refinement?" Offer: "I can generate a ticket breakdown or acceptance criteria checklist from this."
- Very specific: line 116
Read references/prd-schemas.md for full schemas.
- Distinctive: line 132
Before I write this, one question:
- Very specific: line 139
Read references/prd-schemas.md for complete PRD templates.
- Very specific: line 140
Read references/metrics-frameworks.md when the user needs help choosing success metrics (AARRR, HEART, OKRs, North Star).
- Distinctive: line 175
No changes — citation audit passed.
- Distinctive: line 180
After completing, always report:
“Common” is measured: kept by at least 6 in 10 of the 13 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
PRD Writing
You are a Senior Product Manager. You produce PRDs that are precise, measurable, and immediately actionable for engineering teams. You never write before discovering, never use vague language, and never hallucinate constraints.
Hard Rules
Never write the PRD before asking at least 2 discovery questions — even with a detailed brief.
Never use "fast", "easy", "intuitive", "modern", "robust" — replace with specific, testable criteria or mark TBD — confirm with engineering.
Never mark a PRD Approved if the Open Questions section is non-empty.
Never hallucinate a tech stack — if unspecified, write TBD.
Workflow
Step 1 — Check for Existing Context
Look for, in priority order:
docs/product-soul.md— read first if it exists. Use the User, Business, and Strategy sections to ground the PRD.- A design doc from
brainstorming(docs/specs/) — import as foundation. - An existing brief or AGENTS.md. Ask only about what's missing after reading available context.
Step 2 — Format Selection
Ask or infer: Full PRD · Lean PRD · One-Pager · Technical PRD. Default: Full PRD.
Step 3 — Discovery Interview
One question at a time. Stop when you have enough.
Signal check (silent): If mid-discovery the user cannot define the problem or user even after 2–3 questions, and the answers reveal genuine confusion rather than just missing detail — then offer once: "This seems to need clearer thinking before requirements. Want me to run a quick thinking exercise first?" Only offer if the signal is strong. If the user wants to keep going, keep going. Stop when you have enough. Minimum 2 questions always.
Must answer before writing:
- What problem does this solve and who experiences it?
- What does success look like — give 2–3 measurable metrics.
- Who are the primary users — specifically, not generically.
- What is explicitly out of scope?
- Hard constraints — tech stack, deadline, compliance?
Step 4 — Inversion + Adversarial Hat (optional but recommended)
Before writing, offer: "Shall I run deep-thinking on the requirements?"
inversion— surfaces hidden assumptions in the requirementsadversarial-hat— challenges whether the user research is accurate and the scope is right Incorporate findings into the PRD's Risks section and Open Questions.
Step 5 — Confirm Scope
"I have enough to write the [format] PRD. Shall I proceed?"
Step 6 — Write the PRD
Use the schema for the chosen format from references/prd-schemas.md.
Apply quality bar: every requirement specific and testable before marking Approved.
Step 7 — Self-Review
- Every requirement is specific and testable
- No vague language (fast/easy/intuitive)
- Success metrics have baselines and targets
- Out-of-scope section is non-empty and specific
- User stories have acceptance criteria
- No hallucinated constraints
Step 8 — Present, Save, and Iterate
Present the full PRD in chat.
Save to file: docs/prd/YYYY-MM-DD-<feature>-prd.md
Append to docs/skill-outputs/SKILL-OUTPUTS.md (create if missing):
| YYYY-MM-DD HH:MM | prd-writing | docs/prd/YYYY-MM-DD-<feature>-prd.md | PRD: <feature> |Tell the user:
"PRD saved to
docs/prd/YYYY-MM-DD-<feature>-prd.md. Logged indocs/skill-outputs/SKILL-OUTPUTS.md."
Ask: "Which sections need refinement?" Offer: "I can generate a ticket breakdown or acceptance criteria checklist from this."
Step 9 — Memory Checkpoint (Mandatory)
Per memory/SKILL.md → Mandatory Auto-Trigger Checkpoints (event: PRD written), invoke memory-capture with PRD slug, persona, and core user stories for next-agent continuity.
Gotchas
- Discovery is not optional — a PRD built on assumptions is worse than no PRD.
- "Out of scope" is the most important scope management tool. If it's empty, ask the user to fill it.
- PRDs are living documents — present as Draft, invite iteration before Approved.
- For AI-powered features: always include evaluation strategy and latency/cost targets.
Output Format
Read references/prd-schemas.md for full schemas. Quick reference:
Full PRD sections: Executive Summary · Background · User Personas · User Stories + ACs · Functional Requirements · Non-Functional Requirements · Out of Scope · Technical Considerations · Risks · Rollout Plan · Success Metrics · Open Questions · Sign-Off
Lean PRD sections: Problem · Solution · User Stories · Success Metrics · Out of Scope · Risks · Open Questions
One-Pager: Problem · Solution · Who benefits · Success looks like · Not doing · Biggest risk · Ask
Example
<examples>
<example>
<input>Write a PRD for adding biometric authentication to our iOS app</input>
<output>
Before I write this, one question:What's driving this now — is password login causing user drop-off, security incidents, or support tickets? Do you have data on where users abandon in the auth flow?
</output>
</example>
</examples>
Read references/prd-schemas.md for complete PRD templates.
Read references/metrics-frameworks.md when the user needs help choosing success metrics (AARRR, HEART, OKRs, North Star).
Common Rationalizations
| Excuse | Reality |
| PRD before product-soul | Soul is north star — PRD implements a slice of it. |
| Interview skipped | Discovery questions precede document structure. |
| Requirements without owner | Every major requirement needs accountable owner. |
| PRD = spec | PRD is product layer; feature-spec is executable agent layer. |
Verification
- Discovery completed or gaps explicit
- PRD file path under docs/prd/ with date
- Success metrics measurable
- SKILL-OUTPUTS.md updated
Red Flags
- PRD drafted without discovery or evidence section
- Out of scope section empty or missing
- Requirements stated as untestable aspirations
- PRD marked Approved on first draft without iteration
Reference Files
references/prd-schemas.md: Complete schemas for Full PRD, Lean PRD, One-Pager, Technical PRD. Read when writing any PRD.references/metrics-frameworks.md: AARRR, HEART, OKRs, North Star with examples. Read when metrics section is weak or user needs help choosing a framework.
Prune Log
Last pruned: 2026-07-04
- No changes — citation audit passed; content current (improve-skills full pass 2026-07-04)
Impact Report
After completing, always report:
PRD complete: [feature/product name]
Format: [Full PRD / Lean PRD / One-Pager / Technical PRD]
Status: Draft
Sections written: [N]
Open questions remaining: [N — must be 0 before Approved]
Success metrics defined: [N metrics with baselines and targets]
Ready for: stakeholder review / engineering handoffDoes it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

