Skip to content
Back to skills

Doc Tidy

ASecurity

Scan the repo for scattered documentation files, analyze organization issues, and output cleanup recommendations. Analysis only — no changes made.

  • 11 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added May 26, 2026
developmentgogitapidocumentation

Works with

  • api
  • mcp

Security analysis

A100/100

Scanned October 6, 2026

npx -y skills add tykisgod/quick-question --skill doc-tidy --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Doc Tidy?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Doc Tidy
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/tykisgod-doc-tidy/badge)](https://www.skillsdirectory.com/skills/tykisgod-doc-tidy)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
description: "Scan the repo for scattered documentation files, analyze organization issues, and output cleanup recommendations. Analysis only — no changes made."
---

Respond in the user's preferred language (detect from their recent messages, or fall back to the language setting in CLAUDE.md).

## Execution Steps

### 1. Full Scan

Dispatch 3 parallel Explore subagents (model: haiku) to collect a documentation inventory:

**Subagent A — Root + Docs/**
- All .md files in root directory (except CLAUDE.md, AGENTS.md)
- All files and directory structure under Docs/ (recursive)
- For each file: path, first line content, line count

**Subagent B — In-code documentation under Assets/**
- All .md files under Assets/ (recursive)
- Including README, ARCHITECTURE, API_REFERENCE, CHANGELOG, AGENTS.md, etc.
- For each file: path, first line content, line count

**Subagent C — Config & tool documentation**
- .claude/ directory structure
- Documentation under scripts/
- Templates under .github/
- Root config files (.mcp.json, opencode.json, etc.)

### 2. Classify

Categorize all documents by:

| Category | Description |
|----------|-------------|
| **Long-term design** | System architecture, module design, data config (should be centrally managed) |
| **In-code docs** | README, API reference, dev guides (should live with the code) |
| **Branch artifacts** | Review outputs, spec files, timelines (should be archived after merge) |
| **Temporary files** | tmp-*, one-off bug reports, old prompts (should be cleaned or archived) |
| **Project entry points** | CLAUDE.md, AGENTS.md, PR templates (should stay in root) |
| **Tool config** | Skills, hooks, script docs (each has its place, don't move) |

### 3. Find Issues

Check and report:

- **Root pollution**: files in root that shouldn't be there
- **Unorganized accumulation**: directories with >10 files and no subdirectory structure
- **Temporal mixing**: long-term docs and temporary artifacts in the same directory
- **Duplicate docs**: files with highly similar names or content across directories
- **Orphaned docs**: documents referencing deleted modules or outdated code
- **Missing docs**: Service modules with code but no documentation at all

### 4. Output Cleanup Plan

Format:

```
## Current State Summary
- X documentation files across Y locations
- Main issues: 1, 2, 3 (one sentence each)

## Issue List
For each issue: location, description, number of files affected

## Recommended Plan
### Target Directory Structure
(tree diagram)

### Specific Actions
Priority ordered:
1. Delete (temporary files, confirmed unused)
2. Move (archive, categorize)
3. Merge (eliminate duplicates)
4. Fill gaps (important modules missing docs — flag but don't rush)

### Leave Alone
Explicitly list locations that are well-organized and need no changes
```

## Notes

- **Analysis only — do not execute**: output the plan and wait for user confirmation
- Leave alone design docs synced from external sources (e.g., Notion exports; their structure is managed externally), skill files, and in-code READMEs (unless duplicated)
- For docs where staleness is uncertain, mark as "needs confirmation" rather than suggesting deletion

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.