Skip to content
Agentmart
Submit a tool

brainstorm-flow

Run OpenCaw's persistent pre-planning Brainstorm mode for researched app, game, product, feature, or other idea discovery.

Skill by TimothyMeadows

IndexedMIT★ 4
Write a review

Chalk Labs

Need help deploying brainstorm-flow?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/timothymeadows-brainstorm-flow

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Run OpenCaw's persistent pre-planning Brainstorm mode for researched app, game, product, feature, or other idea discovery. Use when the user explicitly says to start or enter Brainstorm mode, asks to continue an active repository-root BRAINSTORM.md, requests the Brainstorm graph or full Markdown, explicitly exits Brainstorm mode, or selects a completed brainstorm element for later planning.

Use cases: documents

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/brainstorm-flow

What's inside

Skill name
brainstorm-flow
Description
Run OpenCaw's persistent pre-planning Brainstorm mode for researched app, game, product, feature, or other idea discovery. Use when the user explicitly says to start or enter Brainstorm mode, asks to continue an active repository-root BRAINSTORM.md, requests the Brainstorm graph or full Markdown, explicitly exits Brainstorm mode, or selects a completed brainstorm element for later planning.
License
MIT
Instructions
47 lines in SKILL.md · view on GitHub

Files (2)

  • SKILL.md
  • openai.yaml · agents/

SKILL.md (first lines)

# Brainstorm Flow

## When to use

Use only after explicit mode activation such as `start Brainstorm mode` or when repository-root `BRAINSTORM.md` already has `Status: active`. An ordinary request that merely uses the verb “brainstorm” does not activate persistent mode.

Brainstorm is an optional discovery stage before planning:

```text
Brainstorm -> Plan -> Task | Goal | Gauntlet
```

## Workflow

1. Resolve the project root and inspect Brainstorm state with `./commands/brainstorm-mode.sh status`.
2. Before activation or resumed idea work, establish exactly three participants: the main agent cast as `computer-science/project-manager` plus two persistent `computer-science/researcher` subagents. If two researcher slots are unavailable, leave durable state unchanged and block idea processing; never substitute sequential or unassisted research.
3. Start with `./commands/brainstorm-mode.sh start` only after the user explicitly activates the mode. On a new session with active state, reconstitute both researchers before processing another idea.
4. Capture each materially distinct user idea as the next unused `IDEA-NNN`. Clarifications refine that element; they do not silently create a second element.
5. Establish baseline understanding through focused questions. Then place the element under the deepest suitable existing `BR-NNN` branch or create the next unused branch with a valid parent.
6. Give both researchers read-only packets with the idea, established understanding, branch context, and research questions:
   - Researcher one investigates the problem, audience, domain, comparable approaches, and precedent.
   - Researcher two investigates feasibility, constraints, dependencies, risks, start conditions, and measurable completeness.
7. Require source URLs for public/current claims and label evidence, inference, disagreements, and unresolved uncertainty. Keep researcher reports ephemeral.
8. Let only the project-manager synthesize and write the element to `BRAINSTORM.md`. Include every required field and subsection, then run `./commands/validate-brainstorm.sh --phase active`.
9. Use `captured`, `clarifying`, `researching`, `plan-ready`, or `parked`. Set plan readiness to `yes` only when every subsection is substantive and research includes at least one HTTP(S) citation.
10. When the user asks to see the brainstorm, run `./commands/show-brainstorm.sh` and present its Mermaid mindmap. If the user explicitly asks for Markdown, run `./commands/show-brainstorm.sh --markdown` and return the entire file.
11. Stop only after explicit user direction with `./commands/brainstorm-mode.sh stop`. This preserves incomplete elements, marks the mode inactive, and regenerates `BRAINSTORM_SUMMARY.md` as the hash-bound branch index.
12. After exit, wait. Start planning only when the user explicitly selects an element; planning may later feed Task, Goal, or Gauntlet flow.

## Output

- Repository-root `BRAINSTORM.md` containing durable mode, session, branch, and full element state.
- Repository-root `BRAINSTORM_SUMMARY.md` generated only on explicit exit as a current hash-bound index.
- A Mermaid mindmap for the default “show brainstorm” request, or verbatim Markdown when explicitly requested.
- A plan-ready element with evidence, start conditions, and a concrete definition of complete.

## Guardrails

- Never create or mutate task, issue, Goal, Gauntlet, implementation, commit, branch-publication, or PR state while Brainstorm is active.
- Never infer activation or deactivation from inactivity, completeness, or an ordinary ideation request.
- Never create task-backed `SUBAGENTS.md` for Brainstorm. The fixed research team is mode-owned and researchers remain read-only.
- Never let researcher subagents write Brainstorm artifacts; the project-manager owns branch placement, reconciliation, and the sole write set.
- Never mark an element plan-ready with pending sections, unsupported current claims, hidden disagreements, or a vague start/finish boundary.
- Never discard, renumber, or reuse branch or element IDs. Preserve incomplete and parked ideas in the full artifact and exit summary.
- Never put credentials, personal data, private-source content, or unsupported legal, financial, or medical conclusions in Brainstorm artifacts.

From TimothyMeadows/OpenCaw, MIT.

More from TimothyMeadows/OpenCaw

Signals

Stars
4
Forks
3
Open issues
2
Last commit (this folder)
Aug 9, 2026 (2 months ago)
Last release
Aug 11, 2026 (2 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from TimothyMeadows

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get brainstorm-flow running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying brainstorm-flow?

CallGet help
Cite this page

brainstorm-flow skill. Agent Mart. https://agentmart.dev/skills/timothymeadows-brainstorm-flow (last updated 2026-10-24)