# Create Token-Efficient LLM Documentation

Long documentation wastes context and slows AI-assisted work. This skill turns project knowledge into concise tables, bullets, examples, and agent-ready structure.

## Install

```bash
npx skillstore add aceek/llm-doc-writer
```

## Metadata

- - Status: approved
- - Slug: aceek-llm-doc-writer
- - Version: 1.0.0
- - Author: Aceek
- - GitHub username: Aceek
- - License: MIT
- - Repository: https://github.com/Aceek/claude-config/tree/master/src/skills/llm-doc-writer
- - Ref: master
- - Supported tools: Claude, Codex, Claude Code
- - Risk level: safe
- - Quality score: 81
- - Quality tier: silver
- - Public page: https://skillstore.pages.dev/skills/aceek-llm-doc-writer
- - Manifest: https://skillstore.pages.dev/api/skills/aceek-llm-doc-writer/manifest

## Capabilities

- Creates compact documentation layouts for CLAUDE.md, README files, and agent instructions.
- Converts prose-heavy sections into tables, bullets, and short examples.
- Provides structure templates for stack, commands, architecture, and conventions.
- Recommends splitting long documents with links to focused detail files.
- Checks for filler words, redundant statements, and verbose narrative sections.

## Use Cases

- Prepare Agent Context: Create a concise CLAUDE.md or agent instruction file for an existing codebase.
- Compress Technical Docs: Rewrite long setup, architecture, or API sections into scannable documentation.
- Standardize Project READMEs: Build consistent README sections for stack, commands, conventions, and references.

## Prompt Templates

### Draft a Basic Agent Doc

```
Create a CLAUDE.md for this project. Include stack, commands, architecture, conventions, and links to detail files.
```

### Rewrite Verbose Documentation

```
Rewrite this documentation for LLM context. Use tables, bullets, and examples. Remove filler and redundant statements.
```

### Design a Documentation Split

```
Review this long project document. Propose a main agent doc plus focused files for architecture, API, and workflows.
```

### Audit for Token Efficiency

```
Audit this documentation for token efficiency. Identify verbose sections, missing commands, unclear conventions, and content to split.
```

## Limitations

- Does not verify that project commands or architecture facts are correct.
- Does not generate diagrams, screenshots, or external assets.
- Does not replace human review for public product documentation.
- Works best when the user provides accurate project details.

## Best Practices

- Provide source facts before asking the skill to write final documentation.
- Keep the main agent document short and link to focused detail files.
- Review generated commands and architecture notes against the actual repository.

## Anti Patterns

- Do not use the skill to invent missing project details.
- Do not keep long narrative paragraphs when a table or list is clearer.
- Do not place every project detail in a single agent context file.

## Security Audit

- - Safe to publish: true
- - Audited at: 2026-06-28T04:42:56.101\+00:00
- - Summary: Static analysis reported external command, network, weak crypto, and reconnaissance patterns. Manual review found only markdown examples, inline code formatting, and ordinary documentation text, with no executable code or malicious intent.

## Stats

- - Views: 215
- - Downloads: 10
- - Favorites: 0
- - Popularity score: 0