{NAME}
Skill0 starsChanged 16 days agoMITView on GitHub
Use when creating, moving, or naming files and folders in a {FRAMEWORK} project. Covers directory layout, naming, module boundaries, config and test placement, and growth from prototype to large.
In this repository7 other files it gives its agents.
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 6 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
Write short imperative sentences.
- Distinctive: line 14
Keep this file under 150 lines.
- Very specific: line 18
Delete this comment block when done.
- Distinctive: line 31
Read the existing tree before creating files.
- Distinctive: line 42
Read references/full-tree.md when you need the expanded tree with every subfolder.
- Distinctive: line 69
Read references/growth-stages.md when deciding whether to restructure.
- Distinctive: line 91
Do not Fill in.
≈ “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
<!--
HOW TO FILL THIS TEMPLATE
1. Replace every {PLACEHOLDER}. {NAME} is the skill name (kebab-case),
{FRAMEWORK} the language or framework, {VERSION} the version covered.
2. Replace each "Fill in:" line with real content, then delete the line.
3. Write short imperative sentences. Every rule must be actionable when an
agent creates a file ("Put X in Y"), not descriptive.
4. Keep this file under 150 lines. Move full trees and long lists to
references/. Link each reference with a one-line note on when to read it.
Optional folders (assets/, examples/, scripts/, evals/) must be
mentioned in this file or removed. See docs/skill-anatomy.md.
5. Delete this comment block when done.
Naming: `<language>-project-structure` or `<framework>-architecture`.
-->{FRAMEWORK} project structure
Target version
Written for {FRAMEWORK} {VERSION} on {DATE}. Fill in: version covered, date, and any version-specific behavior that changes layout.
Follow the repo first
Read the existing tree before creating files. If the project already has conventions, use them, even where they differ from this skill. Fill in: what to check first (config files, existing folders, lint rules).
Default directory layout
{ROOT}/
{folder}/ # one-line rationaleFill in: the tree, with one line of rationale per top-level folder. Read references/full-tree.md when you need the expanded tree with every subfolder.
Naming conventions
- Files: Fill in.
- Exports: Fill in.
- Modules / components: Fill in.
- Routes: Fill in.
- Tests: Fill in.
Dependency direction
Fill in: which layers may import which. State the allowed direction as a rule, for example "Features import shared. Shared never imports features."
Where things live
- Config: Fill in.
- Env vars: Fill in.
- Shared types: Fill in.
- Global state: Fill in.
Tests
Fill in: where tests go (colocated or mirrored tree), file naming pattern, and where fixtures and helpers live.
Structure at three sizes
Read references/growth-stages.md when deciding whether to restructure.
Prototype
Fill in: the minimal layout, when it is enough.
Mid-size
Fill in: the layout to adopt next. Signal to move up: Fill in a concrete, checkable trigger (file count, team size, duplicated code).
Large
Fill in: the layout for scale. Signal to move up: Fill in a concrete trigger.
Grow by moving files, not by rewriting them. Fill in: how to migrate between sizes.
Anti-patterns
Fill in 5 to 10 rules. Read references/anti-patterns.md when reviewing a layout or before moving files.
- Do not Fill in.
- Do not Fill in.
- Do not Fill in.
- Do not Fill in.
- Do not Fill in.
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.