code-brainstorming
Skill0 starsChanged 7 months agoMITView on GitHub
Turn coding ideas into approved designs and planning-ready implementation briefs before coding starts. Use for new features, meaningful behavior changes, ambiguous implementation requests, refactors that change system shape, or early architecture and design discussions. Do not use for tiny obvious edits, isolated typo fixes, trivial null checks, or other straightforward bug fixes that do not need design work.
In this repository6 other files it gives its agents.
- code-debuggingSkill
- code-explain-visuallySkill
- code-reviewingSkill
- code-simplifierSkill
- git-commitSkill
- git-worktreesSkill
The file
We split this skill into the rules it gives an agent
12 rules. About 0 are common, 12 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈12
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 10
Start by understanding the current project context, then ask questions in a numbered list to refine the idea.
- Distinctive: line 14
Do not write code.
- Distinctive: line 15
Do not scaffold files or projects.
- Distinctive: line 16
Do not take implementation action until the design brief has been presented and approved.
- Distinctive: line 22
Read relevant files, docs, types, configs, and recent changes before asking intent questions.
- Distinctive: line 23
Follow existing patterns instead of inventing unrelated ones.
- Distinctive: line 29
Prefer concrete multiple-choice questions when they reduce ambiguity.
- Distinctive: line 30
Prefer open ended question other wise.
- Distinctive: line 31
Do not ask questions that the repository can answer.
- Distinctive: line 37
Keep scope fixed. If the user suggests a new capability, mark it as deferred and return to the current change.
- Distinctive: line 48
YAGNI ruthlessly: - Remove unnecessary features from all designs
- Distinctive: line 52
Question: - Call out assumptions explicitly
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Brainstorming
Help turn ideas into fully formed designs and specs through natural collaborative dialogue.
Start by understanding the current project context, then ask questions in a numbered list to refine the idea. Once you understand what you're building, present the design and get user approval.
Hard Gate
- Do not write code.
- Do not scaffold files or projects.
- Do not take implementation action until the design brief has been presented and approved.
Workflow
1. Explore context first
- Read relevant files, docs, types, configs, and recent changes before asking intent questions.
- Follow existing patterns instead of inventing unrelated ones.
- Assess scope early. If the request spans multiple independent subsystems, say so and help decompose it before refining one slice.
2. Clarify intent
- Focus on purpose, constraints, success criteria, and boundaries the user cares about.
- Prefer concrete multiple-choice questions when they reduce ambiguity.
- Prefer open ended question other wise.
- Do not ask questions that the repository can answer.
3. Identify gray areas
- Convert the request into specific decisions that will change the design.
- Focus on behavior, interfaces, data flow, states, failure handling, and testing expectations as appropriate.
- Keep scope fixed. If the user suggests a new capability, mark it as deferred and return to the current change.
4. Compare approaches
- Present 2-3 viable approaches.
- Lead with the recommended option.
- Explain tradeoffs in terms of codebase fit, complexity, flexibility, and user goals.
Key Principles
- Multiple choice preferred - Easier to answer than open-ended when possible
- YAGNI ruthlessly - Remove unnecessary features from all designs
- Explore alternatives - Propose 2-3 approaches before settling
- Incremental validation - Present design, get approval before moving on
- Be flexible - Go back and clarify when something doesn't make sense
- Question - Call out assumptions explicitly
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

