Imported from z1digitalstudio/gentle-project (
skills/design-research/SKILL.md). Install upstream withnpx skills add z1digitalstudio/gentle-project --skill design-research. Copyright stays with the author.
Design Research
Turn post-kickoff context into usable research and opportunity framing.
This skill exists to separate evidence from assumptions, clarify users and market context, and leave the project ready for prioritization. It should not jump ahead into roadmap or PRD decisions as if research already resolved them.
When to Use
Use this skill when:
- kickoff context is already usable
- the project needs structured research before prioritization
- user, market, competitor, or opportunity signals are still scattered
- the team needs a reliable
product/RESEARCH.mdbefore defining roadmap
Do not use this skill when:
- kickoff framing is still too weak and
discovery-onboardingorkickoff-closeoutshould run first - the work is already at the stage of prioritizing MVP and releases
- the request is really to write PRD or user stories
Required Reading
Read these files before producing output when they exist:
docs/PROJECT-SUMMARY.mdmanagement/RISKS.mdproduct/RESEARCH.mdproduct/artifacts/Opportunity-Brief.mdproduct/artifacts/Kickoff-Closeout-Summary.md
If one or more files are missing, continue and treat that as missing context rather than a blocker.
Minimum Inputs
Require usable project framing from at least one of these:
docs/PROJECT-SUMMARY.mdproduct/artifacts/Opportunity-Brief.mdproduct/artifacts/Kick-off-Meeting-Prep.mdproduct/artifacts/Kickoff-Closeout-Summary.md- kickoff notes or equivalent external context
- other external documentation that already explains the problem, users, and goals
If there is still not enough context to distinguish signals from assumptions, stop and say which earlier skill should run first.
Recommended Inputs
Use these when available:
- existing
management/RISKS.md - research notes or interviews
- benchmark or competitor review
- links to product, prototype, or adjacent solutions
- support, sales, or commercial data
- externally provided
Brand Personalitywhen the project has a design component
Treat Brand Personality as optional external input, not as output of this skill.
Optional Outputs
Default behavior should prioritize product/RESEARCH.md.
Optional artifacts like Discovery Insights, Personas, and User Journeys should be produced only when they add clear value to product or design decisions.
Workflow
1. Confirm that research is the right next step
Verify that:
- kickoff-level context is already usable
- the project still needs structured research before prioritization
- there is enough evidence or context to produce meaningful findings
If not, say which prerequisite is missing.
2. Rebuild the current context
Extract and normalize:
- problem framing
- project objective
- user or stakeholder context
- business and delivery constraints
- already-known risks
- major open questions
3. Separate evidence from assumptions
Explicitly distinguish:
- evidence backed by inputs
- repeated signals that look promising but still need validation
- assumptions that the team is making without enough proof
- contradictions across sources
Do not blur these categories.
4. Produce or update product/RESEARCH.md
Create or update product/RESEARCH.md as the canonical output.
Prefer this structure:
ContextResearch InputsUser SignalsPain PointsMarket / Competitor SignalsEvidence vs AssumptionsOpportunity AreasOpen QuestionsImplications for Roadmap
5. Produce optional working artifacts when they add value
A. Discovery Insights
If the project needs a richer synthesis artifact, read references/discovery-insights-template.md for the canonical template shape.
When produced, product/artifacts/Discovery-Insights.md should capture:
- metadata
- product main goals
- key opportunities
- ideation / how-might-we
- potential solutions explored
- key insights and findings
- feature list proposal
- assumptions, constraints, dependencies, and open questions
B. Personas
Generate product/artifacts/Personas.md only when personas materially improve product or design decisions.
C. User Journeys
Generate product/artifacts/User-Journeys.md only when journey mapping materially improves understanding of flows, friction, or sequencing.
6. Update project memory
When research changes the framing materially, update:
docs/PROJECT-SUMMARY.mdmanagement/RISKS.mdif new uncertainties or discovery risks appear
Output Contract
Your output must include all of the following:
A. Research status
State one of:
Created research draftUpdated existing researchPrepared research for reviewDesign research blocked — earlier discovery context insufficient
B. Canonical document target
State where product/RESEARCH.md should live or confirm that it was updated there.
C. Evidence vs assumptions summary
List:
- strongest evidence
- strongest signals still needing validation
- assumptions that remain unresolved
- contradictions that matter
D. Optional artifacts produced or skipped
State explicitly:
- whether
Discovery Insightswas created - whether
Personaswere created - whether
User Journeyswere created - why each optional artifact was produced or skipped
E. Risks and implications
List:
- new or reframed risks
- major open questions
- implications that should shape
roadmap-definition
F. Next skill
State one of:
Next: roadmap-definitionNext: kickoff-closeoutNext: gather better research evidence
Choose based on the actual state of the project.
G. Human review checklist
End with a short checklist for the responsible person to validate:
- evidence is not being overstated
- assumptions are explicitly labeled
- opportunities are clear enough to prioritize
- optional artifacts were created only when they added value
- roadmap work can now start
Guardrails
Do not:
- invent certainty from weak research
- convert isolated opinions into user truth
- depend on templates outside this skill folder at runtime
- produce
Brand Personalityas a standard output - write the PRD
- write detailed user stories
- treat research synthesis as final prioritization
Success Criteria
The skill succeeds when product/RESEARCH.md exists in usable form, evidence and assumptions are clearly separated, opportunities are visible, and roadmap-definition can start without reinventing discovery context.
