Explore ideas, compare approaches, and produce a validated spec before planning.

Apache-2.0Auto-check passedAgent Workflows

Install Brainstorm

skills CLI
$ npx skills add hashgraph-online/awesome-codex-plugins --skill brainstorm -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install hashgraph-online/awesome-codex-plugins brainstorm --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/hashgraph-online/awesome-codex-plugins.git skills-src && mkdir -p .claude/skills && cp -r skills-src/plugins/yagizdo/quiver/skills/brainstorm .claude/skills/brainstorm && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
brainstorm
GitHub stars
1.2k
Token cost
~2.9k tokens
SKILL.md length
1,493 words
Files
2
Skills in repo
686
Repo updated
First seen
Licence
Apache-2.0

At a glance

Explore ideas, compare approaches, and produce a validated spec before planning.

  • Works in 8 steps: Validate Input → Understand and Assess Complexity → 5 -- Visual Companion Offer → …
  • The user describes what to build
  • SKILL.md covers Step 0 -- Validate Input, Step 1 -- Understand and…, Step 1.5 -- Visual Companion… and Step 2 -- Clarifying Questions, plus 5 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Brainstorm is an agent skill from hashgraph-online/awesome-codex-plugins. Explore ideas, compare approaches, and produce a validated spec before planning. Use when the user describes what to build, asks how to approach something, proposes a new feature or design change, or needs to explore options before implementation.

Its SKILL.md is about 2.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 1 other file (for example `TEST-PLAN.md`).

It sits in Agent Workflows, covering Brainstorming. It works with Git. The repository describes itself as: A curated list of awesome OpenAI Codex / ChatGPT plugins, skills, and resources. The 1 Codex Marketplace. See live plugins at: https://hol.org/plugins/best-codex-plugins. The licence is Apache-2.0.

When your agent uses it

  • The user describes what to build
  • Asks how to approach something
  • Proposes a new feature
  • Needs to explore options before implementation

Example prompts

  • “/brainstorm”

Workflow steps

8 steps, taken from the step headings in SKILL.md.

  1. Validate Input
  2. Understand and Assess Complexity
  3. 5 -- Visual Companion Offer
  4. Clarifying Questions
  5. Generate Approaches
  6. Executive Summary Gate
  7. Write Spec Document
  8. User Review Gate

What it can do on your machine

Read from SKILL.md and the folder at commit 78497e5. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Brainstorm loads about 2.9k tokens when it runs. Until then it costs about 65 tokens; SKILL.md has 1,493 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~65
When it runs · the whole SKILL.md, loaded when a task matches
~2.9k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from hashgraph-online/awesome-codex-plugins at commit 78497e5, republished under its Apache-2.0 licence (© hashgraph-online). 1,493 words, ~2,900 tokens.

Download SKILL.mdSave it as .claude/skills/brainstorm/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
brainstorm
description
Explore ideas, compare approaches, and produce a validated spec before planning. Use when the user describes what to build, asks how to approach something, proposes a new feature or design change, or needs to explore options before implementation.
argument-hint
<idea or feature description>
when-to-use
user wants to explore ideas, compare approaches, or decide what to build -- '/brainstorm', 'brainstorm this', 'help me think through' (not: 'make a plan')

Gather Context

!`git rev-parse --is-inside-work-tree 2>/dev/null || echo "NO_GIT"`
!`git branch --show-current 2>/dev/null || echo "NO_GIT"`
!`git log --oneline -5 2>/dev/null || echo "NO_COMMITS"`

Instructions

Before starting Step 0, use the Glob tool to gather project context silently (do not show results to the user):

  1. docs/brainstorms/*.md -- existing brainstorm specs
  2. .claude/plans/*.md -- existing plans
  3. *.md and *.json and *.yaml and *.yml in project root -- root config files
  4. src/** or lib/** or app/** or packages/** or commands/** or components/** -- source directory structure (first level only)

Treat empty Glob results as "directory does not exist". Proceed regardless.

You are a brainstorming partner. Your job is to transform vague ideas into validated specs through collaborative dialogue -- asking the right questions, generating concrete approaches with trade-offs, and producing a spec document that is ready for /plan. You do NOT write code or implementation plans -- you explore, clarify, and document the design.

Step 0 -- Validate Input

If any gather-context block returned NO_GIT, this directory is not a git repository. Print: > No git repository detected -- skipping branch/commit context. Proceed to Step 1. Treat all git-sourced fields as empty.

If $ARGUMENTS is empty and the conversation has no obvious pending idea:

No idea to brainstorm. Usage: /brainstorm <describe your idea or challenge> Stop here.

Step 1 -- Understand and Assess Complexity

Restate the idea in one sentence, then assess its complexity silently (do not show the assessment label to the user).

DepthSignalsBehavior
Quick1-3 files, single layer, well-understood domain, clear scope0-1 clarifying questions, 2 short approaches, brief spec (~50 lines)
Standard3-10 files, 2+ layers, some ambiguity or design choices2-3 clarifying questions, 2-3 detailed approaches, standard spec (~100-150 lines)
Deep10+ files, architectural impact, new domain, security/auth/payments3-4 clarifying questions, 3 comprehensive approaches with alternatives considered, full spec (~200+ lines)

Quick-exit for trivial ideas: If the idea is trivial (single file, obvious implementation), use AskUserQuestion:

This is straightforward enough to plan directly without a brainstorm session. Buttons: ["Skip brainstorm -- go to /plan", "Brainstorm anyway"]

If user picks "Skip brainstorm", stop here and suggest running /plan <task>.

Decomposition Check

If the description contains 3+ independent subsystems, spans multiple technology layers (backend + frontend + mobile + infra), or estimates 20+ affected files, use AskUserQuestion:

This idea contains multiple independent subsystems. Splitting into sub-projects produces better specs than cramming everything into one. Buttons: ["Split into sub-projects", "Continue as single spec"]

If "Split": List the identified sub-projects with a recommended order. Then continue the normal brainstorm flow (Step 2 onward) for the first sub-project only. At the end (Step 7), note: "Remaining sub-projects: [list]. Run /brainstorm for each when ready."

If "Single spec": Continue normal flow. User decision takes priority.

Step 1.5 -- Visual Companion Offer

If the idea involves UI/UX, layout, design, or architecture diagrams, use AskUserQuestion:

This topic is well-suited for visual content. I can open a browser-based companion to show mockups and diagrams as we go. Want to try it? Buttons: ["Yes, open visual companion", "No, continue with text"]

If "Yes": Read the visual-companion skill for setup instructions. Start the companion server and keep it running throughout the session; write HTML content for visual questions and continue in the terminal for text-only ones. If "No": Continue with normal text-only flow. Skip this step entirely for pure backend, data model, or API-design topics.

Step 2 -- Clarifying Questions

Ask questions to fill gaps in your understanding. Focus on: purpose, constraints, success criteria, existing patterns to follow or break.

Rules:

  • Use AskUserQuestion with action buttons -- never ask as plain text.
  • Derive button options from the idea context and codebase -- not generic placeholders.
  • Always include an "Other" free-text option as the last button (AskUserQuestion adds this automatically).
  • Question count follows depth: Quick = 0-1, Standard = 2-3, Deep = 3-4.
  • Group independent questions. If two questions have no dependency on each other, ask them in the same AskUserQuestion call using the multi-question format (up to 4 questions per call). Only separate questions when the answer to one determines what you ask next.
  • After each answer, decide: enough context to proceed, or one more question needed? Do not ask questions for the sake of filling a quota. Skip entirely if the user provided clear scope/constraints/success criteria, the idea is a well-known pattern (CRUD, auth, endpoint), or the user explicitly said "just brainstorm approaches".

Step 3 -- Generate Approaches

Present 2-3 distinct approaches. Each approach must include:

  1. Name -- short, descriptive (e.g., "Event-driven pipeline", "Simple polling loop")
  2. How it works -- 3-5 sentences describing the approach
  3. Trade-offs -- explicit pros and cons
  4. Best for -- when this approach shines

Lead with your recommendation. Mark it clearly and explain why in 1-2 sentences. Do not be neutral -- take a position.

Rules:

  • Approaches must be genuinely different strategies grounded in actual project structure and conventions from gather-context. Penalize complexity that does not serve stated requirements.
  • If the idea has a "standard way" in the project's stack, include it as one approach even if you recommend something different.
  • For Quick depth: 2 approaches, concise (3-4 lines each). For Deep: 3 approaches, detailed (8-12 lines each).

Present approaches and use AskUserQuestion:

Which approach do you want to go with? Buttons: approach names as labels, one-line summaries as descriptions.

Step 4 -- Executive Summary Gate

After the user selects an approach, present a concise executive summary. This is the primary approval gate -- the user should be able to approve or reject without reading the full spec.

## Executive Summary

**What:** [1 sentence -- what we are building/changing]
**Approach:** [selected approach name] -- [1 sentence why]
**Key decisions:**
- [decision 1]
- [decision 2]
- [decision 3 if needed]

**Scope:**
- Touches: [modules/areas affected]
- Out of scope: [what we are NOT doing]

Use AskUserQuestion:

Does this direction look right? Buttons: ["Approve -- write the spec", "Adjust -- I want to change something", "Restart -- pick a different approach"]

Handle each response:

  • Approve -- proceed to Step 5.
  • Adjust -- ask what to change, revise the summary, and re-present.
  • Restart -- go back to Step 3 with the adjustment context.
Show full SKILL.md (570 more words)Show less

Step 5 -- Write Spec Document

Write the validated design as a spec document. Scale section depth to complexity -- a Quick spec can be 50 lines, a Deep spec can be 200+. Not every section is required for every depth.

Required sections by depth:

  • Quick: Executive Summary, Design (Chosen Approach + Key Decisions + Affected Areas), Success Criteria. Omit Context, Alternatives, Open Questions unless they add clear value.
  • Standard: All sections above plus Context and Open Questions. Skip Alternatives Considered.
  • Deep: All sections including Alternatives Considered and Design Principles Applied.

Section content rules:

  • Header block: # [Feature/Idea Name] -- Brainstorm Spec with Date, Status: Draft, Depth fields.
  • Executive Summary: copy the approved summary from Step 4.
  • Design > Chosen Approach: detailed description of the selected approach.
  • Design > Key Decisions: numbered list with brief rationale per decision.
  • Design > Affected Areas: file paths, modules, or system boundaries.
  • Alternatives Considered (Deep only): other approaches from Step 3 with rejection reasons.
  • Open Questions: items deferred to planning phase.
  • Success Criteria: how we know it is done and working.
  • Design Principles Applied (Standard + Deep): YAGNI (features deferred and why), Single Responsibility (each unit's role), Interface Clarity (communication interfaces).
  • No placeholders -- every included section must have real content. Remove sections you cannot fill.
  • Always write in English unless the user explicitly requests another language.
Save the spec:
  1. Create docs/brainstorms/ if it does not exist.
  2. Write the spec as docs/brainstorms/YYYY-MM-DD-<descriptive-name>.md (use date '+%Y-%m-%d' for the date prefix).
  3. Verify: Read the file back and confirm it was written correctly. Then self-review with fresh eyes: (1) placeholder scan -- any "TBD", "TODO", or vague requirements? Fix inline; (2) internal consistency -- do sections contradict each other or diverge from the executive summary? Fix inline; (3) scope check -- is this focused enough for a single /plan session, or should it be decomposed into sub-specs? (4) ambiguity check -- can any requirement be read two ways? Pick one interpretation and make it explicit. Fix any issues by editing the saved file.

Step 7 -- User Review Gate

Use AskUserQuestion:

Spec saved. You can review it now or move forward. What would you like to do? Buttons: ["Looks good -- move to planning", "Let me review first", "Save and stop"]

  • Looks good -- move to planning -- invoke the plan skill with the spec path as context.
  • Let me review first -- say: "Take your time. Let me know if you want changes or if we should move to /plan." Stop here and wait.
  • Save and stop -- stop here.

Anti-Patterns

  • Don't skip the executive summary gate -- it exists so users do not have to read 200-line specs to approve direction.
  • Don't write implementation details or code -- that is /plan's job. Specs describe WHAT and WHY, not HOW at the code level.
  • Don't ask clarifying questions as plain text -- always use AskUserQuestion with action buttons.
  • Don't present more than 3 approaches -- decision fatigue kills momentum. 2-3 is the sweet spot.
  • Don't be neutral on approaches -- always lead with a recommendation and defend it.
  • Don't force questions when the user gave a clear, detailed description -- assess and skip if appropriate.
  • Don't show depth labels ("This is Standard depth") to the user -- depth is internal routing logic.
  • Don't leave placeholder sections in the spec ("TBD", "TODO") -- either fill them or remove them.
  • Don't add features the user didn't ask for and don't architect for hypothetical future requirements -- solve the problem at hand. If "we might need X later" comes up, note it in Open Questions only.

© hashgraph-online, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in plugins/yagizdo/quiver/skills/brainstorm of hashgraph-online/awesome-codex-plugins.

  • SKILL.md
  • TEST-PLAN.md

Open the folder on GitHubat commit 78497e5

Compare with similar skills

Brainstorm next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Brainstorm compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Brainstorm this skillhashgraph-online/awesome-codex-plugins1.2k—~2.9kAutomated safety check: PassApache-2.0
Trellis StartROYIANS/foliq-print-template-designer1356 repos~646Automated safety check: PassMIT
Callabracadabra50/claude-code-voice-skill172—~759Automated safety check: NotesMIT
Consult Codextobihagemann/turbo407—~4.1kAutomated safety check: PassMIT
Pm Brainstormbex-co/beancount-io295—~1.6kAutomated safety check: PassMIT
Brainstormingxpinjection/test-driven-spring-boot11254 repos~2.6kAutomated safety check: PassMIT

Similar skills

  • Trellis Start

    ROYIANS/foliq-print-template-designer

    Initializes an AI development session by reading workflow guides, developer identity, git status, active tasks, and project guidelines from .trellis/.

    135 GitHub starsUsed in 6 repos~646 tokens
    Agent WorkflowsAuto-check passed
  • Call

    abracadabra50/claude-code-voice-skill

    Voice calls with Claude about your coding projects via Vapi.

    172 GitHub stars~759 tokensUpdated 2 mo ago
    Agent WorkflowsAuto-check: notes
  • Consult Codex

    tobihagemann/turbo

    Multi-turn consultation with Codex CLI for second opinions, brainstorming, or collaborative problem-solving.

    407 GitHub stars~4.1k tokensUpdated yesterday
    Agent WorkflowsAuto-check passed
  • Pm Brainstorm

    bex-co/beancount-io

    Propose milestones and tasks for the repository's public .pm adoption board — decompose a topic through the TPM adoption lens, size and gate the work, and emit a text-only proposal for /pm to…

    295 GitHub stars~1.6k tokensUpdated yesterday
    Agent WorkflowsAuto-check passed
  • Brainstorming

    xpinjection/test-driven-spring-boot

    You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior.

    112 GitHub starsUsed in 54 repos~2.6k tokens
    Agent WorkflowsAuto-check passed
  • Darwin Skill Optimizer

    alchaincyf/darwin-skill

    Scores SKILL.md files on a nine-dimension rubric, then improves them in a keep-or-revert loop with independent judge agents, test prompts, git history and human checkpoints.

    6.2k GitHub starsUsed in 1 repo~4.7k tokens
    Agent WorkflowsAuto-check passed

More from hashgraph-online/awesome-codex-plugins

All 686 skills in this repo
  • Anime Reaction Gif

    hashgraph-online/awesome-codex-plugins

    Create original anime-style reaction stickers as looping GIFs and MP4 previews, using generated character pose sheets and timed key poses.

    1.2k GitHub stars~922 tokensUpdated today
    Auto-check passed
  • Calibredb

    hashgraph-online/awesome-codex-plugins

    Manage and query Calibre libraries with the calibredb CLI (local paths or Calibre Content server URLs).

    1.2k GitHub stars~1k tokensUpdated today
    Auto-check passed
  • Rust API Test Harness

    hashgraph-online/awesome-codex-plugins

    A skill your agent uses when adding, changing, testing, or debugging Rust HTTP APIs and services, especially when Codex needs black-box integration tests, random-port app startup, real database test…

    1.2k GitHub stars~1.7k tokensUpdated today
    Auto-check passed
  • Art

    hashgraph-online/awesome-codex-plugins

    Make a studio's game look like something at build time — a cover from a real frame of the game (free), painted covers, backdrops, textures and character plates from image models through the…

    1.2k GitHub stars~2.4k tokensUpdated today
    Auto-check passed
  • Game Balance Economy

    hashgraph-online/awesome-codex-plugins

    Balance game difficulty, resources, rewards, probability, progression, economies, and dominant strategies.

    1.2k GitHub stars~618 tokensUpdated today
    Auto-check passed
  • Manuscript Engagement Analytics

    hashgraph-online/awesome-codex-plugins

    Analyze nonfiction manuscripts for reader engagement signals, including heading-level word counts, slow starts, long slogs, weak takeaway titles, value pacing, beta-reader comment dropoff, and…

    1.2k GitHub stars~875 tokensUpdated today
    Auto-check passed

Works with

Categories

Questions about Brainstorm

What does Brainstorm do?

Explore ideas, compare approaches, and produce a validated spec before planning. Brainstorm is an agent skill from hashgraph-online/awesome-codex-plugins. Explore ideas, compare approaches, and produce a validated spec before planning.

When should I use Brainstorm?

Brainstorm fits situations like: the user describes what to build; asks how to approach something; proposes a new feature; needs to explore options before implementation.

How do I install Brainstorm in Claude Code?

Run `npx skills add hashgraph-online/awesome-codex-plugins --skill brainstorm -a claude-code`. Or copy the skill folder (plugins/yagizdo/quiver/skills/brainstorm in hashgraph-online/awesome-codex-plugins) into .claude/skills/brainstorm in your project. Claude Code loads it when a task matches its description.

How do I install Brainstorm in Codex?

Run `npx skills add hashgraph-online/awesome-codex-plugins --skill brainstorm -a codex`. Or copy the skill folder (plugins/yagizdo/quiver/skills/brainstorm in hashgraph-online/awesome-codex-plugins) into .agents/skills/brainstorm in your project. Codex loads it when a task matches its description.

Can I use Brainstorm in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add hashgraph-online/awesome-codex-plugins --skill brainstorm -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/brainstorm, .gemini/skills/brainstorm, .github/skills/brainstorm and .opencode/skills/brainstorm in your project.

What does Brainstorm need to run?

SKILL.md names no scripts, command-line tools or credentials: Brainstorm is instructions for the agent only.

Does Brainstorm access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Brainstorm safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Brainstorm use?

Brainstorm is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Brainstorm use?

About 2.9k tokens (SKILL.md is roughly 12k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Brainstorm?

Skills that share tags, products or a category with Brainstorm: Trellis Start (ROYIANS/foliq-print-template-designer, 135 stars), Call (abracadabra50/claude-code-voice-skill, 172 stars), Consult Codex (tobihagemann/turbo, 407 stars) and Pm Brainstorm (bex-co/beancount-io, 295 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Brainstorm?

hashgraph-online (a GitHub organization) maintains it in hashgraph-online/awesome-codex-plugins, which has 1,242 GitHub stars. The repository holds 686 skills in this directory. The repository was last updated on October 8, 2026.

Source: hashgraph-online/awesome-codex-plugins on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.