plugins/workflow/skills/brainstorming/SKILL.md
Use when the user wants to explore approaches or discuss design before implementation. Explores intent, requirements, and design through collaborative dialogue.
npx skillsauth add rbergman/dark-matter-marketplace brainstormingInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Turn ideas into fully formed designs through natural collaborative dialogue.
Process: Understand context → Ask questions one at a time → Explore approaches → Present design incrementally → Validate each section.
Before asking questions:
Rules:
Focus on:
Before exploring approaches, run these four checks on the idea:
| Check | Question | Red Flag | |-------|----------|----------| | Clarity | Can someone explain what this does in one sentence? | They describe internals, not outcomes | | Timeline | Is the value immediate or does it compound over time? | Neither — value is vague or hypothetical | | Perception | Can the user see/feel it working? | Value is invisible or requires explanation | | Discovery | Does the user seek this out, or must they stumble into it? | Must be explained repeatedly to new users |
Any red flags → address them as design constraints in Phase 2, not afterthoughts.
Once you understand the goal:
Example:
I see three approaches:
1. **Component-based** (recommended) - Fits your existing pattern,
easiest to test, but more files.
2. **Inline** - Fewer files, but harder to reuse and test.
3. **Hook-based** - Most flexible, but adds complexity you may not need.
I'd go with #1 because [reasoning]. Thoughts?
Once approach is chosen:
Cover:
Write validated design to:
docs/plans/YYYY-MM-DD-<topic>-design.md
Commit to git.
Ask: "Ready to set up for implementation?"
Then:
dm-work:worktrees to create isolated workspacedm-work:orchestrator patterns for delegation| Principle | Why | |-----------|-----| | One question at a time | Don't overwhelm | | Multiple choice preferred | Easier to answer | | YAGNI ruthlessly | Remove unnecessary features | | Explore alternatives | Always 2-3 approaches before settling | | Incremental validation | Present design in sections | | Be flexible | Go back when something doesn't fit |
| Don't | Do Instead | |-------|------------| | Jump to implementation | Understand first, design second | | Ask 5 questions at once | One question per message | | Present monolithic design | Break into 200-300 word sections | | Skip trade-off discussion | Always propose 2-3 approaches | | Assume you understand | Validate understanding with user |
1. Check project context (files, docs, commits)
2. Ask questions one at a time (prefer multiple choice)
3. Propose 2-3 approaches with trade-offs
4. Present design in sections, validate each
5. Write to docs/plans/YYYY-MM-DD-<topic>-design.md
6. If implementing: worktree → beads → orchestrate
development
Initialize a new repository with standard scaffolding - git, gitignore, AGENTS.md, justfile, mise, beads, and timbers. Use when starting a new project or setting up an existing repo for Claude Code workflows.
data-ai
Activate at session start when using Agent Teams for complex multi-agent work. Establishes team lead role with delegation protocols, teammate spawning, model selection, and beads integration. You coordinate the team; teammates implement.
data-ai
Use when creating a worktree, setting up a worktree, starting feature work that needs isolation, or before executing implementation plans. Covers git worktree creation under .worktrees/, gitignore setup, beads integration, and merge guardrails.
data-ai
Activate when you are a delegated subagent (not the orchestrator). Establishes subagent protocol with terse returns, details to history/, file ownership boundaries, and escalation rules. You implement; orchestrator reviews and commits.