design-brief-generator
Skill8 starsChanged 28 days agoMITView on GitHub
Run upstream design discovery and produce a design brief: problem (5 Ws), users and tasks, business value, Lean UX hypothesis and riskiest assumption, scope, constraints (including accessibility target), risks, and success metrics, with every gap marked TBD. Two modes: guided discovery (asks one section at a time) or synthesis (turns intake-form answers, Slack threads or kickoff notes into a brief). Also produces an async intake form. Use when the user says "write a design brief", "project kickoff", "discovery questions", "intake form for this request", "what problem are we solving", "brief from these notes", or is starting a UX project before a solution is chosen. Not for a build-ready feature spec or acceptance criteria (use intent-spec or spec-pack), and not for capturing what a sync decided vs. left open (use requirements-to-decision-doc).
2 other projects use a version of this.
Adopted by
Most recently taken up by
- Samueladewole/claude-skillstheir own version, first seen Oct 13, 2026
- jamesrochabrun/skillstheir own version, first seen Nov 18, 2025
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- jamesrochabrun/skills★ 209same name
- Samueladewole/claude-skills★ 0same name
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 repository19 other files it gives its agents.
- brand-tokens-translatorSkill
- case-study-captureSkill
- component-generatorSkill
- content-writerSkill
- design-accessibilitySkill
- design-lead-thought-partnerSkill
- flow-captureSkill
- flow-diagram-renderSkill
- icon-generatorSkill
- intent-specSkill
- meeting-actions-to-tasksSkill
- portable-prototypeSkill
- requirements-to-decision-docSkill
- responsive-auditSkill
- spec-packSkill
- theme-generatorSkill
- token-auditSkill
- ui-edge-case-auditSkill
- visual-qaSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 5 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Run upstream design discovery and produce a structured design brief before anyone commits to a solution.
- Distinctive: line 34
Ask one section per turn, at most four questions.
- Distinctive: line 37
If Constraints is skipped, still record the accessibility target as WCAG 2.2 AA (assumed), so it reaches the brief.
- Distinctive: line 48
Use the 5 Ws to define the problem clearly:
- Very specific: line 115
After gathering information, generate the brief using the template in references/prd-template.md.
- Distinctive: line 123
If information is missing:
- Very specific: line 125
Add [TBD] placeholders with specific questions to resolve
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Design Brief Generator
Run upstream design discovery and produce a structured design brief before anyone commits to a solution. Two modes: guided discovery (interactive Q&A) or synthesis (process existing inputs).
Where this fits
The brief captures why, for whom and what success looks like, and marks every gap TBD. It is deliberately not a build spec. When the team is ready to build, hand off:
- Build-ready feature spec and acceptance criteria →
intent-specorspec-pack - What a meeting decided vs. left open →
requirements-to-decision-doc
Workflow
- Determine mode:
- Guided discovery? → Walk through the question set, one section per turn (see Pacing)
- Synthesis mode? → User provides existing inputs (form responses, notes, Slack threads) → Extract and organize into the brief, then ask only about required sections that are still empty
- Async intake form requested? → Read
references/intake-form.mdand produce the form
- Gather information using the question set below
- Generate the brief using the template in
references/prd-template.md - Offer the hand-off — name the spec skill that comes next
Pacing (guided mode)
The full set is about 40 questions. Asked all at once, people skim and answer only the easy ones, so:
- Ask one section per turn, at most four questions. Split a longer section (Constraints) across two turns.
- Required: Problem Statement, Users & Tasks, Scope, Success. Everything else is optional: offer it in one line ("Want to cover stakeholders and risks now, or mark them TBD?") and move on if declined.
- Skip questions already answered. When you can infer an answer, state it and let the user correct it rather than asking again.
- If Constraints is skipped, still record the accessibility target as WCAG 2.2 AA (assumed), so it reaches the brief.
Question Set
Work through these sections. Skip questions already answered. Flag gaps.
Project Basics
- Project/feature name — What are we calling this?
- Trigger — What initiated this? (Customer pain, compliance change, internal initiative, leadership request, tech debt?)
Problem Statement
Use the 5 Ws to define the problem clearly:
- Who is experiencing this problem?
- What is the problem or unmet need?
- Where does it occur? (Context, touchpoint, platform)
- When does it happen? (Frequency, conditions)
- Why does it matter? (Impact on user, impact on business)
Then synthesize into a user need statement:
"As a [user type], I need [need/goal] so that [reason/outcome]."
Users & Tasks
- Primary users — Who is this for? (Customers, admins, internal staff, partners, support agents?)
- User goal — What are they trying to accomplish?
- Primary tasks — What specific workflows or routine tasks does the user perform to achieve this goal? List the key steps.
- Pain points — What's broken or frustrating about their current experience?
Business Value
- Business goal — What business outcome does this support? (Revenue, efficiency, compliance, retention?)
- Why now — Why is this a priority right now?
- Value if solved — What's the impact on the business if we solve this well?
Stakeholders & Ownership
- Requestor — Who asked for this?
- Decision-maker — Who has final sign-off?
- Team — Who's involved? (PM, eng lead, other designers, compliance?)
Hypotheses & Assumptions
Frame beliefs using the Lean UX hypothesis format:
"We believe [this business outcome] will be achieved if [this user] attains [this benefit] with [this feature/solution]."
- Key assumptions — What must be true for this to work?
- Riskiest assumption — Which assumption, if wrong, would cause the entire idea to fail?
Scope
- Must-haves — What does the solution absolutely need to do?
- Nice-to-haves — What would be great but isn't critical?
- Out of scope — What are we explicitly not solving?
- Existing patterns — Any features or flows we should reference or extend?
Constraints
- Accessibility target — WCAG 2.2 AA unless the user names another. Public-sector work can add a legal standard: Section 508 for US federal agencies (it incorporates WCAG 2.0 AA) or ADA Title II for US state and local governments (WCAG 2.1 AA). Ask which applies, and whether any known users rely on assistive technology.
- Compliance/regulatory — Legal or regulatory requirements that apply?
- Only if rules differ by region: which jurisdictions or markets are in scope?
- Technology — What's the current tech stack? Any limitations or integration requirements?
- Platform — Web, mobile, both? Browser requirements? Offline needs?
- Timeline — When does this need to ship? Hard deadlines?
- Resources — Budget or staffing constraints?
Risks
- Known risks — What could go wrong? (Design, technical, adoption, compliance?)
- Dependencies — What else needs to happen for this to succeed? What's on the critical path?
- Mitigation — How might we reduce the biggest risks?
Success
- Definition of done — How do we know this is "finished"? (Include the accessibility target.)
- Success metrics — What will we measure? (Adoption, task completion, error reduction, support tickets?)
- Baseline — Current state we're improving from? (Include numbers if available)
- Target — What does success look like quantitatively?
Context & Inputs
- Existing research — Data, user feedback, or insights available?
- Comparables — Competitors or products that do this well?
- Previous attempts — Has this been tried before? What happened?
- Links — Relevant docs, Jira tickets, Figma files, Slack threads?
Output
After gathering information, generate the brief using the template in references/prd-template.md. The brief serves:
- Stakeholder alignment — A shareable artifact for a PM/eng/design sync
- Gap identification — Missing information flagged explicitly
- Hand-off — Seed criteria and a short "For the build agent" note, so the spec stage starts from the brief instead of re-asking
Handling Incomplete Information
If information is missing:
- Generate what you can
- Add
[TBD]placeholders with specific questions to resolve - Include a "Gaps to Resolve" section at the end listing open questions
Reference Files
references/prd-template.md— The brief template. Read it when generating the brief.references/intake-form.md— Async intake form fields. Read it only when the user asks for a form to send out.
Does 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.

