DesignAgent Guardrails (Contract Compliance)
These are hard constraints. Do NOT violate:
- Do not skip any step in contract.json entry.steps
- Do not merge or reorder output sections
- Do not invent data or hallucinate facts
- Do not execute outside the defined step order
- If input is incomplete, ask only for missing required fields
name: 02-discover
description: "Use after intake — research users, context, competitors, and constraints. First divergence phase. Do NOT skip to solutions."
02-Discover: Research and Define
Overview
Second step: diverge into research. Understand users, context, competitors, and constraints, then converge into a written design brief. This combines research (what was formerly design-discover) with synthesis into a problem definition (what was formerly design-define). The goal is insight first, then a clear brief.
Non-Negotiable Rule
HARD GATE: Complete research synthesis AND a written design brief before generating any concepts. "I already know the problem" is not research. A design brief must exist before any visual exploration begins.
When To Use
- After intake is complete (01-intake done)
- Before any concept generation or visual exploration
When NOT To Use
- During intake (you don't know enough yet — finish 01 first)
- When the user asks for a quick sketch (route to Lite mode)
- When all the research already exists and just needs to be referenced (document it and move to 03)
Process
Phase A: Research (Diverge)
1. Stakeholder and User Research
- Who are the primary and secondary users?
- What are their goals, pain points, and behaviors?
- Context of use: where, when, how, with what devices/space?
- Current workflow or experience (if redesign)
2. Context and Competitive Audit
- Existing brand assets, design system, guidelines
- Competitive landscape: direct, aspirational, cross-industry
- Cultural and market context
- Platform / medium constraints
3. Gather Inspiration and References
- Collect visual references, precedents, case studies
- If visual research is needed, invoke visual-research
- Document what works and what doesn't
Phase B: Define (Converge)
4. Synthesize Research
- Cluster findings into themes
- Identify contradictions and tensions
- Document unknowns requiring validation
- Produce a one-paragraph research summary
5. Write the Design Brief
- Problem statement (what are we solving?)
- Project scope and constraints
- Target audience and key insights
- Success criteria (from intake, refined)
- Design principles (3-5 guiding rules)
- Must include: what is in scope AND what is out of scope
Rationalization Prevention
| Excuse |
Reality |
| "This is a small project, I don't need research" |
Every design lives in a context. Skipping research = designing blind. |
| "I already know this industry" |
Every project is unique. Assume nothing. |
| "Let me sketch while I research" |
Sketches without research are decoration, not design. |
| "The brief is obvious" |
Write it down anyway. Obvious briefs hide the most assumptions. |
Red Flags
- You are generating visual ideas before the brief is written
- You have not asked the user a clarifying question
- The "brief" is a paragraph with no constraints or out-of-scope section
- You skipped competitive context
Verification
→ Continue to: use 03-strategy to set direction before generating concepts.
1---2name: 02-discover3description: DesignAgent Guardrails (Contract Compliance)4---5## DesignAgent Guardrails (Contract Compliance)67These are hard constraints. Do NOT violate:89- Do not skip any step in contract.json entry.steps10- Do not merge or reorder output sections11- Do not invent data or hallucinate facts12- Do not execute outside the defined step order13- If input is incomplete, ask only for missing required fields1415---16name: 02-discover17description: "Use after intake — research users, context, competitors, and constraints. First divergence phase. Do NOT skip to solutions."18---1920# 02-Discover: Research and Define2122## Overview23Second step: diverge into research. Understand users, context, competitors, and constraints, then converge into a written design brief. This combines research (what was formerly design-discover) with synthesis into a problem definition (what was formerly design-define). The goal is insight first, then a clear brief.2425## Non-Negotiable Rule26**HARD GATE: Complete research synthesis AND a written design brief before generating any concepts.** "I already know the problem" is not research. A design brief must exist before any visual exploration begins.2728## When To Use29- After intake is complete (01-intake done)30- Before any concept generation or visual exploration3132## When NOT To Use33- During intake (you don't know enough yet — finish 01 first)34- When the user asks for a quick sketch (route to Lite mode)35- When all the research already exists and just needs to be referenced (document it and move to 03)3637## Process3839### Phase A: Research (Diverge)4041#### 1. Stakeholder and User Research42- Who are the primary and secondary users?43- What are their goals, pain points, and behaviors?44- Context of use: where, when, how, with what devices/space?45- Current workflow or experience (if redesign)4647#### 2. Context and Competitive Audit48- Existing brand assets, design system, guidelines49- Competitive landscape: direct, aspirational, cross-industry50- Cultural and market context51- Platform / medium constraints5253#### 3. Gather Inspiration and References54- Collect visual references, precedents, case studies55- If visual research is needed, invoke **visual-research**56- Document what works and what doesn't5758### Phase B: Define (Converge)5960#### 4. Synthesize Research61- Cluster findings into themes62- Identify contradictions and tensions63- Document unknowns requiring validation64- Produce a one-paragraph research summary6566#### 5. Write the Design Brief67- Problem statement (what are we solving?)68- Project scope and constraints69- Target audience and key insights70- Success criteria (from intake, refined)71- Design principles (3-5 guiding rules)72- Must include: what is in scope AND what is out of scope7374## Rationalization Prevention7576| Excuse | Reality |77|--------|---------|78| "This is a small project, I don't need research" | Every design lives in a context. Skipping research = designing blind. |79| "I already know this industry" | Every project is unique. Assume nothing. |80| "Let me sketch while I research" | Sketches without research are decoration, not design. |81| "The brief is obvious" | Write it down anyway. Obvious briefs hide the most assumptions. |8283## Red Flags84- You are generating visual ideas before the brief is written85- You have not asked the user a clarifying question86- The "brief" is a paragraph with no constraints or out-of-scope section87- You skipped competitive context8889## Verification90- [ ] Research summary produced (one paragraph minimum)91- [ ] At least 3 user needs / pain points documented92- [ ] Competitive context mapped (at least 2 peers)93- [ ] Design brief written with problem statement, scope, audience, criteria, principles94- [ ] In-scope AND out-of-scope explicitly listed95- [ ] No design solutions have been generated yet9697→ Continue to: use **03-strategy** to set direction before generating concepts.