From space-agents
Produce structured spec.md from brainstorm conversations. Creates a consistent format that feeds cleanly into the planning phase.
npx claudepluginhub thebrownproject/space-agents --plugin space-agentsThis skill uses the workspace's default tool permissions.
Transform brainstorm conversations into structured spec documents. This skill produces consistent `spec.md` files that the planning phase can reliably parse.
Creates isolated Git worktrees for feature branches with prioritized directory selection, gitignore safety checks, auto project setup for Node/Python/Rust/Go, and baseline verification.
Executes implementation plans in current session by dispatching fresh subagents per independent task, with two-stage reviews: spec compliance then code quality.
Dispatches parallel agents to independently tackle 2+ tasks like separate test failures or subsystems without shared state or dependencies.
Transform brainstorm conversations into structured spec documents. This skill produces consistent spec.md files that the planning phase can reliably parse.
Called at the end of /brainstorm when the user confirms they're ready to capture their exploration. Can also be invoked directly if you have a clear idea ready to document.
Location: .space-agents/exploration/ideas/YYYY-MM-DD-<topic>/spec.md
The spec file lives alongside any exploration notes in the same topic folder. This folder will later move to planned/ when /plan processes it.
# Exploration: [Topic Title]
**Date:** YYYY-MM-DD
**Status:** [Ready for planning | Needs discussion | Blocked]
---
## Problem
What problem are we solving? Why does this matter?
[1-3 paragraphs describing the problem, its impact, and why it needs solving]
---
## Solution
High-level approach. The "what" not the "how".
[Describe the solution at a conceptual level. Include diagrams/tables if they clarify the approach.]
---
## Requirements
Must-have functionality. Use checklist format for tracking.
- [ ] Requirement 1
- [ ] Requirement 2
- [ ] Requirement 3
---
## Non-Requirements
Explicitly out of scope. Prevents scope creep during planning.
- Not doing X
- Not changing Y
- Not supporting Z
---
## Architecture
Components, data flow, key technical decisions.
[Include diagrams, flow descriptions, component interactions. This section should give a developer enough context to understand how the pieces fit together.]
---
## Constraints
Technical limitations, patterns to follow, dependencies.
- [Constraint 1 - e.g., must work with existing API]
- [Constraint 2 - e.g., follow project naming conventions]
- [Constraint 3 - e.g., external dependency requirement]
---
## Success Criteria
How do we know it's done? Checklist format for verification.
- [ ] Criterion 1
- [ ] Criterion 2
- [ ] Criterion 3
---
## Open Questions
Things still to be decided. Unresolved questions may block planning.
1. [Question 1] - [Context/options if known]
2. [Question 2] - [Context/options if known]
---
## Next Steps
What happens after this exploration.
1. `/plan` to create implementation tasks
2. [Any specific preparatory work needed]
3. [Dependencies to resolve first]
| Section | Purpose | Planning Impact |
|---|---|---|
| Problem | Why are we doing this? | Guides priority decisions |
| Solution | What approach? | Shapes feature structure |
| Requirements | What must it do? | Becomes task acceptance criteria |
| Non-Requirements | What's out of scope? | Prevents scope creep |
| Architecture | How do pieces fit? | Informs task breakdown |
| Constraints | What limits us? | Affects implementation choices |
| Success Criteria | How do we verify? | Defines "done" for feature |
| Open Questions | What's unresolved? | Blocks planning if critical |
| Next Steps | What's next? | Guides user action |
/plan/brainstorm conversation
|
v
/exploration-write-spec creates:
exploration/ideas/YYYY-MM-DD-<topic>/spec.md
|
v
/plan reads spec.md, creates plan.md, moves to:
exploration/planned/YYYY-MM-DD-<topic>/
|
v
/plan creates Beads, moves to:
mission/staged/YYYY-MM-DD-<topic>/