organize
Skill4 starsChanged 4 months agoMITView on GitHub
Repository structure only: directory layout, file placement, naming conventions, and where-does-this-belong decisions. Detects the project's convention, audits files against it, and executes move plans with import-path updates. Never changes code inside files beyond the import updates a move forces; in-file restructuring is /refactor.
14 other projects use a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- paulogabriel/beryltheir own version, first seen Nov 8, 2026
- SethGammon/Citadeledited, first seen Nov 4, 2026
- yuryalencar/ai-skillstheir own version, first seen Nov 1, 2026
- zarubinvibe/athenatheir own version, first seen Nov 1, 2026
- snuffish/dotfilestheir own version, first seen Nov 1, 2026
- mcpware/cross-code-organizertheir own version, first seen Oct 25, 2026
Other versions14 projects use a version of this.
Word for word · 0
None found.
Edited · 1
- SethGammon/Citadel★ 92798% same
Wrote their own version · 13
- yunshu0909/yunshu_skillshub★ 760same name
- mcpware/cross-code-organizer★ 385same name
- mcpware/claude-code-organizer★ 383same name
- TyrealQ/q-skills★ 108same name
- sanghun0724/cmux-claude-skills★ 33same name
- seongjaeryu/bindly-claude-code★ 8same name
- zarubinvibe/athena★ 2same name
- GRIDLOCK-NYC/claude-skills★ 0same name
- ianpugfaced55/claude-code-organizer★ 0same name
- paulogabriel/beryl★ 0same name
3 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository50 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- architectSkill
- archonSkill
- ascii-diagramSkill
- autopilotSkill
- costSkill
- create-appSkill
- create-skillSkill
- daemonSkill
- dashboardSkill
- decision-mapSkill
- deploy-stewardSkill
- designSkill
- doc-genSkill
- doSkill
- evolveSkill
- experimentSkill
- fleetSkill
- grillSkill
- housecleanSkill
- improveSkill
- infra-auditSkill
- learnSkill
- live-previewSkill
- loopSkill
- mapSkill
- marshalSkill
- merge-reviewSkill
- postmortemSkill
- prdSkill
- pr-watchSkill
- qaSkill
- refactorSkill
- research-fleetSkill
- researchSkill
- reviewSkill
- scaffoldSkill
- scheduleSkill
- session-handoffSkill
- setupSkill
- systematic-debuggingSkill
- telemetrySkill
- test-genSkill
- triageSkill
- unharnessSkill
- verifySkill
- watchSkill
- wikiSkill
- workspaceSkill
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 1 is distinctive, and 3 are very specific to this project.
0
Common
Most of the 11 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 54
Read .claude/harness.json.
- Very specific: line 58
Use the Glob tool (not find or Get-ChildItem) to discover directories.
- Distinctive: line 71
Check every source file against the placement rules (manifest if present, detected convention otherwise).
- Very specific: line 116
Conflict with
protectedFiles: if protected files conflict with placement rules, warn and ask which takes precedence.
“Common” is measured: kept by at least 6 in 10 of the 11 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/organize -- Repository Structure
Orientation
Use when: Deciding where a file belongs, auditing file placement against the project's conventions, running a naming-convention sweep, or moving files so the tree matches the declared layout.
Don't use when: You want a structural map of code relationships (use /map), you want to change code inside files (use /refactor), you need disk or artifact cleanup (use /houseclean), or you need a single known move executed (use /refactor move).
Boundary with /refactor
/organize changes where files live and what they are named. Changing code structure within files (extract, inline, split functions or components, rename symbols) is /refactor. The only in-file edits /organize makes are import-path updates forced by a move or rename. If an audit reveals a file that should be split or merged, recommend /refactor; do not do it here.
Commands
| Command | Behavior |
/organize | Detect convention, audit placement, propose and execute a move plan |
/organize --audit | Check current files against the manifest, report violations only |
/organize --show | Display the current organization manifest |
/organize --lock | Set locked: true so the enforcement hook blocks violations |
/organize --unlock | Set locked: false so enforcement is advisory |
/organize where does {file} go | Single placement decision, no full audit |
Protocol
Step 1: CHECK
Read .claude/harness.json. If an organization key exists, use its convention and placement rules; on bare /organize, ask whether to audit, adjust the rules, or reconfigure. If no key exists, continue to Step 2.
Step 2: DETECT
Use the Glob tool (not find or Get-ChildItem) to discover directories. Filter out node_modules, .git, .planning, .citadel, .claude, dist, build, __pycache__, .next, target, .venv, venv. Cap at 200 directories; if larger, scan only the top 3 levels.
Convention signals:
src/components/,src/hooks/,src/utils/-> layer-basedsrc/features/auth/,src/features/dashboard/-> feature-basedsrc/auth/components/,src/auth/hooks/-> hybrid- Flat
src/with no subdirectories -> flat - Mixed signals -> custom (ask the user which convention to adopt)
Record: convention, confidence (high/medium/low), roots with purposes, anomalies.
Step 3: AUDIT
Check every source file against the placement rules (manifest if present, detected convention otherwise). Also check naming: files or directories that break the dominant casing or suffix pattern of their siblings.
=== Structure Audit: {project} ===
Convention: {convention} ({confidence})
Compliant: {N}/{M} source files
Violations:
{current_path} -> {expected_path} ({rule broken})
Naming:
{path}: {issue}Step 4: PLAN
For each violation, build a move plan before touching anything: source path -> destination path, every importer whose import path must change, and barrel/index files that need re-exports updated. Moves and renames only; never delete files with content. Present the plan and wait for confirmation.
Step 5: APPLY
Execute confirmed moves, update all import paths referencing moved files, update barrels, then run the project's typecheck command (from harness.json). Zero new errors allowed. If a move fails typecheck for reasons beyond import paths, revert that single move and flag the file instead of editing its code.
Step 6: CONFIGURE
On first run or reconfigure, write the organization key to harness.json. Read-modify-write: preserve all other keys, and preserve any existing dynamic/cleanupPolicy entries this skill no longer manages.
{
"organization": {
"convention": "layer",
"roots": { "src": { "purpose": "application source" } },
"placement": [
{ "glob": "*.test.{ts,tsx}", "rule": "colocated", "target": null, "reason": "tests live next to source" }
],
"locked": false
}
}Rule types: colocated, sibling-dir, root-dir, within-root. Set locked: false initially; /organize --lock turns on blocking enforcement via the organize-enforce hook.
Fringe Cases
- No directories found / empty project: suggest flat convention; nothing to audit.
- Monorepo (multiple package.json): scan each package root separately; ask whether rules are per-package or repo-wide.
- harness.json missing: audit in advisory mode from the detected convention; offer to create the manifest.
- User changes convention: warn with the number of files that would move. Never mass-move without explicit confirmation.
- Conflict with
protectedFiles: if protected files conflict with placement rules, warn and ask which takes precedence.
Contextual Gates
Disclosure: "Auditing repository structure. Moves execute only after you confirm the plan."
Reversibility: amber -- file moves and import-path updates; undo with git checkout -- . before commit or git revert after.
Trust gates: Any: detect and audit. Familiar (5+ sessions): confirmed move plans execute autonomously.
Quality Gates
- Convention detected or read from the manifest before any audit verdicts
- Every violation lists current path, expected path, and the rule it breaks
- Move plan presented and confirmed before any file moves
- No in-file code edits beyond import-path updates; split/extract needs are deferred to /refactor
- Typecheck clean after moves (zero new errors)
- harness.json write preserves all keys outside
organization
Exit Protocol
---HANDOFF---
- Structure: {convention}; {compliant}/{total} files compliant after changes
- Moves: {N} files moved, {M} import paths updated, 0 in-file refactors
- Deferred to /refactor: {split/extract recommendations, or "none"}
- Enforcement: {"advisory (unlocked)" | "blocking (locked)"}
- Reversibility: amber -- undo moves with git checkout/revert
---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.

