doc-generate
Skill0 starsChanged 8 months agoMITView on GitHub
Generate structured project documentation: architecture overview, API reference, or developer onboarding guide. Scans the codebase and produces markdown files in docs/. Triggers: "generate docs", "documentation", "onboarding guide", "document API", "architecture doc".
- Reads credentials
2 other projects use a version of this.
Adopted by
Most recently taken up by
- allanninal/claude-code-skillstheir own version, first seen Oct 17, 2026
- ryusuke-ai/argustheir own version, first seen Feb 13, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- allanninal/claude-code-skills★ 1same name
- ryusuke-ai/argus★ 1same name
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 repository20 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- ci-watchSkill
- commitSkill
- discovery-interview-prepSkill
- incident-responseSkill
- opportunity-solution-treeSkill
- polarisSkill
- prd-developmentSkill
- prioritization-advisorSkill
- reviewSkill
- roadmapSkill
- security-auditSkill
- session-wrapSkill
- spec-updateSkill
- tddSkill
- tech-debt-auditSkill
- test-runnerSkill
- user-storySkill
- uxSkill
- workshop-facilitationSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
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 24
Ask the user which documentation to generate:
- Distinctive: line 52
Test commands and CI configuration
- Very specific: line 57
Write markdown files to docs/ directory:
- Distinctive: line 194
Ask the user to review for accuracy
- Distinctive: line 206
Generating docs nobody reads: Keep docs short and scannable.
≈ “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
Documentation Generator — Workflow
When to use
- New team member joining -- Generate an onboarding guide so they can ramp up fast
- After a major feature or refactor -- Update architecture docs to reflect current state
- API consumers need reference -- Generate endpoint documentation from code
- Project has no documentation -- Bootstrap docs/ from scratch
- Never for trivial updates (single function change, typo fix)
Process
Step 1: Detect documentation mode
Ask the user which documentation to generate:
- Architecture -- System overview, components, data flow, patterns, Mermaid diagrams
- API -- REST/GraphQL endpoints with methods, parameters, responses, examples
- Onboarding -- Developer setup guide, conventions, workflow, codebase tour
- All -- Generate all three (architecture first, then API, then onboarding)
Step 2: Scan the codebase
Based on the selected mode, scan relevant files:
Architecture mode:
- Project structure (directories, key files)
- Entry points (main, index, app files)
- Configuration files (package.json, docker-compose, CI config)
- Module boundaries and imports graph
- Design patterns in use (MVC, hexagonal, event-driven, etc.)
API mode:
- Route definitions (Express routes, FastAPI endpoints, controller files)
- Request/response schemas (DTOs, types, validation)
- Middleware chain (auth, validation, error handling)
- OpenAPI/Swagger specs if present
Onboarding mode:
- README.md (existing setup instructions)
- CLAUDE.md (conventions and workflow)
- Package manager and dependency files
- Test commands and CI configuration
- Environment variables (.env.example)
Step 3: Generate documentation
Write markdown files to docs/ directory:
Architecture output: docs/architecture.md
# Architecture — [Project Name]
## Overview
[1-2 paragraph summary of what the system does and how it's structured]
## System Diagram
```mermaid
graph TD
A[Client] --> B[API Gateway]
B --> C[Service Layer]
C --> D[Database]Components
[Component Name]
- Purpose: [what it does]
- Location:
path/to/module/ - Key files:
file1.ts,file2.ts - Dependencies: [what it imports]
- Dependents: [what imports it]
Data Flow
[Describe how data moves through the system]
Design Patterns
- [Pattern name]: [where and why it's used]
Key Decisions
- [Decision 1 and rationale]
#### API output: `docs/api.md`
```markdown
# API Reference — [Project Name]
## Base URL
`{{BASE_URL}}`
## Authentication
[Auth method and usage]
## Endpoints
### [Resource Name]
#### `GET /api/resource`
**Description:** [what it does]
**Parameters:**
| Name | Type | Required | Description |
| --- | --- | --- | --- |
| limit | number | No | Max results (default: 20) |
**Response (200):**
```json
{
"data": [],
"total": 0
}Errors:
| Code | Description |
| 401 | Unauthorized |
| 404 | Resource not found |
#### Onboarding output: `docs/onboarding.md`
```markdown
# Developer Onboarding — [Project Name]
## Prerequisites
- [Runtime version]
- [Package manager]
- [Required tools]
## Setup
### 1. Clone and install
```bash
git clone [repo-url]
cd [project]
[install command]2. Environment configuration
cp .env.example .env
# Fill in: [list of required variables]3. Run the project
[start command]4. Run tests
[test command]Project Structure
project/
├── src/ → [description]
├── tests/ → [description]
└── docs/ → [description]Conventions
- [Coding convention 1]
- [Commit format]
- [Testing approach]
Common Tasks
- Add a new feature: [workflow]
- Fix a bug: [workflow]
- Deploy: [workflow]
Architecture Overview
See architecture.md for the full system design.
### Step 4: Review with user
After generating docs:
- Present a summary of what was generated
- Ask the user to review for accuracy
- Iterate on any corrections or additions
- Suggest committing with `/commit`
## After generation
- **Commit the docs** with `/commit` when the user approves
- **Keep docs updated** -- re-run after major changes
- **Link from README** -- add links to generated docs in the project README
## Common Pitfalls
- **Generating docs nobody reads:** Keep docs short and scannable. If a section needs more than 2 scrolls, split it.
- **Documenting implementation details:** Focus on "what" and "why", not "how" at the line level. Code is the source of truth for implementation.
- **Stale docs:** Generated docs drift from code. Re-run periodically or after major changes.
- **Missing the audience:** Onboarding is for new devs, architecture is for the team, API is for consumers. Don't mix audiences.
- **Over-diagramming:** One clear diagram beats five complex ones. Use Mermaid for the main system view, not every function call.
## References
### External Frameworks
- Daniele Procida, *Diataxis Documentation Framework* (2017) -- tutorials, how-to, reference, explanation
- Simon Brown, *C4 Model for Software Architecture* (2006) -- context, container, component, code diagrams
- Tom Preston-Werner, *Readme Driven Development* (2010)
### Related Skills
- `/spec-update` -- Generates functional specs per domain scope (invoked by /commit step 6)
- `/review` -- May flag missing documentation as a suggestion
- `/tech-debt-audit` -- Includes documentation gaps in its assessment
### Credit
- Inspired by Desktop Commander "Create Project Documentation" and "Generate Dev Onboarding Guide" prompt patterns.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.

