Brainstorm - Design-First Ideation
Scheduling
Goal
Explore user intent, constraints, and alternative approaches before planning or implementation, then preserve an approved design for downstream planning.
Intent signature
- User says they have an idea, want to brainstorm, compare approaches, explore concepts, or design before planning.
- Request is ambiguous enough that implementation or task planning would be premature.
When to use
- Exploring a new feature idea before planning
- Understanding user intent and constraints before committing to an approach
- Comparing multiple design approaches with trade-offs
- When the user says "I have an idea" or "let's design something"
- Before invoking
/plan for complex or ambiguous requests
When NOT to use
- Requirements are already clear and well-defined -> use
oma-pm directly
- Implementing actual code -> delegate to specialized agents
- Performing code reviews -> use
oma-qa
- Debugging existing issues -> use
oma-debug
Expected inputs
- Early idea, ambiguous goal, product concept, design question, or set of constraints
- Existing project context when the idea must fit a codebase or product direction
- User preferences and approval gates
Expected outputs
- Clarified intent and constraints
- Two or three approaches with tradeoffs and a recommended option
- Section-by-section approved design document
- Saved design artifact before handoff to planning
Dependencies
- Shared context loading, reasoning templates, clarification protocol, quality principles, and skill routing
- Downstream PM workflow for task decomposition after design approval
Control-flow features
- Branches by ambiguity, user answers, approach comparison, and approval gates
- Asks one question at a time
- Stops before implementation or task planning
Structural Flow
Entry
- Confirm that the request is exploratory rather than ready for implementation.
- Load enough project context to understand constraints.
- Start with intent and constraints, not solutions.
Scenes
- PREPARE: Explore context and frame the design question.
- ACQUIRE: Ask clarifying questions one at a time.
- REASON: Generate two or three approaches with tradeoffs.
- VERIFY: Get user approval section by section.
- FINALIZE: Save design and transition to planning when appropriate.
Transitions
- If requirements become clear and implementation-ready, transition to PM planning.
- If user rejects an approach, revise before moving to detailed design.
- If implementation pressure appears early, defer it until design approval.
Failure and recovery
- If the user cannot answer a question, propose assumptions and ask for confirmation.
- If scope expands, split the design into smaller sections.
- If alternatives collapse into one option, identify the real constraint causing that.
Exit
- Success: approved design exists and is ready for planning.
- Partial success: open questions and assumptions are explicit.
Logical Operations
Actions
| Action |
SSL primitive |
Evidence |
| Read context and idea |
READ |
User prompt and project context |
| Ask targeted questions |
REQUEST |
Clarification phase |
| Compare approaches |
COMPARE |
Tradeoff matrix |
| Infer recommendation |
INFER |
Recommended option |
| Validate approval |
VALIDATE |
Section-by-section confirmation |
| Write design artifact |
WRITE |
docs/plans/designs/ and memory |
| Transition to plan |
NOTIFY |
Handoff summary |
Tools and instruments
- Context loading, reasoning templates, clarification protocol
- Project memory and
docs/plans/designs/ for persisted designs
Canonical workflow path
1. Ask one clarifying question at a time.
2. Present 2-3 approaches with tradeoffs and a recommended option.
3. Save the approved design to `docs/plans/designs/` before handing off to planning.
Resource scope
| Scope |
Resource target |
MEMORY |
User intent, assumptions, decisions |
CODEBASE |
Existing project context when relevant |
LOCAL_FS |
Approved design artifacts |
Preconditions
- The user is still exploring or the request is ambiguous.
- The agent can ask clarifying questions before implementation.
Effects and side effects
- Produces design decisions and persisted design docs.
- Influences downstream planning but does not implement code.
Guardrails
- No implementation or planning before design approval - brainstorm produces a design document, not code or task plans
- One question at a time - ask clarifying questions sequentially, not in batches
- Always propose 2-3 approaches - include a recommended option with trade-off analysis
- Section-by-section design - present design incrementally with user confirmation at each step
- YAGNI - do not over-engineer; design only what is needed for the stated goal
- Save design, then transition - persist the approved design document before handing off to
/plan
Execution Phases
Follow the brainstorm workflow step by step:
- Phase 1 - Context: Explore the existing codebase and understand the project landscape
- Phase 2 - Questions: Ask clarifying questions one at a time to understand intent and constraints
- Phase 3 - Approaches: Propose 2-3 approaches with a recommended option and trade-off matrix
- Phase 4 - Design: Present the detailed design section by section, getting user approval at each step
- Phase 5 - Documentation: Save the approved design to
docs/plans/designs/ and project memory
- Phase 6 - Transition: Hand off to
/plan for task decomposition
Common Pitfalls
- Jumping to solutions: Asking "how" before fully understanding "what" and "why"
- Too many questions at once: Overwhelming the user with a wall of questions
- Single approach bias: Presenting only one option without alternatives
- Over-engineering: Designing for hypothetical future requirements instead of stated needs
- Skipping confirmation: Moving forward without explicit user approval on design decisions
References
Vendor-specific execution protocols are injected automatically by oma agent:spawn.
Source files live under ../_shared/runtime/execution-protocols/{vendor}.md.
- Context loading:
../_shared/core/context-loading.md
- Reasoning templates:
../_shared/core/reasoning-templates.md
- Clarification protocol:
../_shared/core/clarification-protocol.md
- Quality principles:
../_shared/core/quality-principles.md
- Skill-to-agent mapping:
../_shared/core/skill-routing.md
1---2name: oma-brainstorm3description: Design-first ideation that explores user intent, constraints, and approaches before any planning or implementation. Use for brainstorming, ideation, exploring concepts, and evaluating approaches.4---5
6# Brainstorm - Design-First Ideation
7
8## Scheduling
9
10### Goal
11Explore user intent, constraints, and alternative approaches before planning or implementation, then preserve an approved design for downstream planning.
12
13### Intent signature
14- User says they have an idea, want to brainstorm, compare approaches, explore concepts, or design before planning.
15- Request is ambiguous enough that implementation or task planning would be premature.
16
17### When to use
18- Exploring a new feature idea before planning
19- Understanding user intent and constraints before committing to an approach
20- Comparing multiple design approaches with trade-offs
21- When the user says "I have an idea" or "let's design something"
22- Before invoking `/plan` for complex or ambiguous requests
23
24### When NOT to use
25- Requirements are already clear and well-defined -> use `oma-pm` directly
26- Implementing actual code -> delegate to specialized agents
27- Performing code reviews -> use `oma-qa`
28- Debugging existing issues -> use `oma-debug`
29
30### Expected inputs
31- Early idea, ambiguous goal, product concept, design question, or set of constraints
32- Existing project context when the idea must fit a codebase or product direction
33- User preferences and approval gates
34
35### Expected outputs
36- Clarified intent and constraints
37- Two or three approaches with tradeoffs and a recommended option
38- Section-by-section approved design document
39- Saved design artifact before handoff to planning
40
41### Dependencies
42- Shared context loading, reasoning templates, clarification protocol, quality principles, and skill routing
43- Downstream PM workflow for task decomposition after design approval
44
45### Control-flow features
46- Branches by ambiguity, user answers, approach comparison, and approval gates
47- Asks one question at a time
48- Stops before implementation or task planning
49
50## Structural Flow
51
52### Entry
531. Confirm that the request is exploratory rather than ready for implementation.
542. Load enough project context to understand constraints.
553. Start with intent and constraints, not solutions.
56
57### Scenes
581. **PREPARE**: Explore context and frame the design question.
592. **ACQUIRE**: Ask clarifying questions one at a time.
603. **REASON**: Generate two or three approaches with tradeoffs.
614. **VERIFY**: Get user approval section by section.
625. **FINALIZE**: Save design and transition to planning when appropriate.
63
64### Transitions
65- If requirements become clear and implementation-ready, transition to PM planning.
66- If user rejects an approach, revise before moving to detailed design.
67- If implementation pressure appears early, defer it until design approval.
68
69### Failure and recovery
70- If the user cannot answer a question, propose assumptions and ask for confirmation.
71- If scope expands, split the design into smaller sections.
72- If alternatives collapse into one option, identify the real constraint causing that.
73
74### Exit
75- Success: approved design exists and is ready for planning.
76- Partial success: open questions and assumptions are explicit.
77
78## Logical Operations
79
80### Actions
81| Action | SSL primitive | Evidence |
82|--------|---------------|----------|
83| Read context and idea | `READ` | User prompt and project context |
84| Ask targeted questions | `REQUEST` | Clarification phase |
85| Compare approaches | `COMPARE` | Tradeoff matrix |
86| Infer recommendation | `INFER` | Recommended option |
87| Validate approval | `VALIDATE` | Section-by-section confirmation |
88| Write design artifact | `WRITE` | `docs/plans/designs/` and memory |
89| Transition to plan | `NOTIFY` | Handoff summary |
90
91### Tools and instruments
92- Context loading, reasoning templates, clarification protocol
93- Project memory and `docs/plans/designs/` for persisted designs
94
95### Canonical workflow path
96```text
971. Ask one clarifying question at a time.
982. Present 2-3 approaches with tradeoffs and a recommended option.
993. Save the approved design to `docs/plans/designs/` before handing off to planning.
100```
101
102### Resource scope
103| Scope | Resource target |
104|-------|-----------------|
105| `MEMORY` | User intent, assumptions, decisions |
106| `CODEBASE` | Existing project context when relevant |
107| `LOCAL_FS` | Approved design artifacts |
108
109### Preconditions
110- The user is still exploring or the request is ambiguous.
111- The agent can ask clarifying questions before implementation.
112
113### Effects and side effects
114- Produces design decisions and persisted design docs.
115- Influences downstream planning but does not implement code.
116
117### Guardrails
1181. **No implementation or planning before design approval** - brainstorm produces a design document, not code or task plans
1192. **One question at a time** - ask clarifying questions sequentially, not in batches
1203. **Always propose 2-3 approaches** - include a recommended option with trade-off analysis
1214. **Section-by-section design** - present design incrementally with user confirmation at each step
1225. **YAGNI** - do not over-engineer; design only what is needed for the stated goal
1236. **Save design, then transition** - persist the approved design document before handing off to `/plan`
124
125### Execution Phases
126Follow the brainstorm workflow step by step:
1271. **Phase 1 - Context**: Explore the existing codebase and understand the project landscape
1282. **Phase 2 - Questions**: Ask clarifying questions one at a time to understand intent and constraints
1293. **Phase 3 - Approaches**: Propose 2-3 approaches with a recommended option and trade-off matrix
1304. **Phase 4 - Design**: Present the detailed design section by section, getting user approval at each step
1315. **Phase 5 - Documentation**: Save the approved design to `docs/plans/designs/` and project memory
1326. **Phase 6 - Transition**: Hand off to `/plan` for task decomposition
133
134### Common Pitfalls
135- **Jumping to solutions**: Asking "how" before fully understanding "what" and "why"
136- **Too many questions at once**: Overwhelming the user with a wall of questions
137- **Single approach bias**: Presenting only one option without alternatives
138- **Over-engineering**: Designing for hypothetical future requirements instead of stated needs
139- **Skipping confirmation**: Moving forward without explicit user approval on design decisions
140
141## References
142Vendor-specific execution protocols are injected automatically by `oma agent:spawn`.
143Source files live under `../_shared/runtime/execution-protocols/{vendor}.md`.
144- Context loading: `../_shared/core/context-loading.md`
145- Reasoning templates: `../_shared/core/reasoning-templates.md`
146- Clarification protocol: `../_shared/core/clarification-protocol.md`
147- Quality principles: `../_shared/core/quality-principles.md`
148- Skill-to-agent mapping: `../_shared/core/skill-routing.md`