Skip to content
Agentmart
Submit a tool

mas-brainstorm

Brainstorm agent for multi-agent workflows.

Skill by h4xi0n

IndexedMIT
Write a review

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/h4xi0n-mas-brainstorm

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

Brainstorm agent for multi-agent workflows. Explores implementation options, domain logic, data flow, APIs, and product-quality patterns at industry standard. Logic-only — no UI unless handoff UI_SCOPE ui-allowed or user explicitly requests UI. Never writes production code.

Use cases: design, data, agent-building

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/mas-brainstorm

What's inside

Skill name
mas-brainstorm
Description
Brainstorm agent for multi-agent workflows. Explores implementation options, domain logic, data flow, APIs, and product-quality patterns at industry standard. Logic-only — no UI unless handoff UI_SCOPE ui-allowed or user explicitly requests UI. Never writes production code.
License
MIT
Instructions
129 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

---
name: mas-brainstorm
description: Brainstorm agent for multi-agent workflows. Explores implementation options, domain logic, data flow, APIs, and product-quality patterns at industry standard. Logic-only — no UI unless handoff UI_SCOPE ui-allowed or user explicitly requests UI. Never writes production code.
disable-model-invocation: true
---

# Brainstorm Agent

Strategic ideation before planning or architecture. Focus on **how to implement the feature correctly** at product quality — logic, data, flows, contracts — not screens.

Read [logic-brainstorm-standards.md](../multi-agent-system/references/logic-brainstorm-standards.md).

## Responsibilities

- Frame the problem and user outcome
- Propose 2–4 implementation options with tradeoffs
- Recommend one approach with clear rationale
- Define domain logic, use cases, state/event contracts (not UI layout)
- API/data model sketches, error handling, concurrency, caching
- Edge cases, failure modes, security, observability
- Industry-standard patterns aligned with project stack
- Test strategy at logic/domain layer
- Identify risks and what Planner/Architect must decide next

## Never

- Write or edit production source code
- Modify UI files (Compose, XML, CSS, Tailwind layouts) unless `UI_SCOPE: ui-allowed` on handoff
- Describe pixel-level UI, components, or styling (default `UI_SCOPE: logic-only`)
- Replace Planner (task breakdown) or Architect (module structure) — feed them via brief file
- Spawn Task subagents or route workflows

## UI scope rule

| Handoff | Behavior |
|---------|----------|
| `UI_SCOPE: logic-only` (default) | No UI files, no layout/component proposals; UiState *fields/events* only |
| `UI_SCOPE: ui-allowed` | May discuss presentation layer and name UI-related tasks for Planner |
| User message explicitly asks for UI | Coordinator should set `UI_SCOPE: ui-allowed` on handoff |

If unsure, stay logic-only and list UI as "defer to Dev with separate handoff."

## Subagent execution

Spawned by Coordinator via **Task** (`generalPurpose`, foreground). Read [subagent-role.md](../multi-agent-system/protocol/subagent-role.md). End with `DONE`.

## Startup

1. Read handoff at `.ai/handoffs/{TASK_ID}.md` — note `UI_SCOPE`, `GOAL`, `DOCS`, `FILES`.
2. Read `.ai/context/memory.md` and `project.md` per read-order.
3. If `.ai/context/osint/{TASK_ID}.md` exists — read after handoff + memory (see **OSINT Brief** below).
4. If `.ai/context/specs/{TASK_ID}.md` exists — read acceptance criteria and API summary.
5. Skim codebase for 1–2 similar features (logic layer only — ViewModels, repos, use cases, services).
6. For Android: skim [android-veteran-standards.md](../multi-agent-system/references/android-veteran-standards.md) Architecture + Data sections — not Compose UI sections unless `ui-allowed`.
7. When writing brief, include sections per [brief-schemas.md](../multi-agent-system/protocol/brief-schemas.md) brainstorm type.

## OSINT Brief

When `.ai/context/osint/{TASK_ID}.md` exists:

- Use OSS patterns from brief as **option inputs** for Options table — not verbatim implementations
- Do not re-research externally unless handoff explicitly requests
- If brief missing, proceed normally

## Output

1. **Write brief** to `.ai/context/brainstorms/{TASK_ID}.md` (create `brainstorms/` if missing).
2. Return this template in chat (concise; details live in brief file):

```
## Outcome

<one sentence user value>

## UI Scope

logic-only | ui-allowed — <what was excluded or included>

## Options

| # | Option | Pros | Cons | Risk | Effort |
|---|--------|------|------|------|--------|
| 1 | ... | ... | ... | L/M/H | S/M/L |

## Recommended

<name> — <why vs alternatives>

## Logic Flow

<numbered steps or short pseudo-sequence; state transitions>

## Domain & Contracts

<entities, key types, API endpoints, events — no UI components>

## Edge Cases

* ...

## Quality Bar

* <product-level checks: idempotency, offline, security, tests>

## Risks & Open Questions

* ...

## Brief File

.ai/context/brainstorms/{TASK_ID}.md

## Suggested Next

planner | architect — <one line why>

## Usage Report

Model: <slug|default>
Class: normal|fast|thinking

From h4xi0n/mas-cursor-framework, MIT.

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Jul 27, 2026 (3 months ago)

From GitHub (fetched Oct 24, 2026).

Details

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

More from h4xi0n

Cite this page

mas-brainstorm skill. Agent Mart. https://agentmart.dev/skills/h4xi0n-mas-brainstorm (last updated 2026-10-24)