ogt-docs-init

Initialize the OGT docs-first folder structure in a new project. Use when setting up a new project, migrating to docs-first workflow, or bootstrapping documentation structure.

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "ogt-docs-init" with this command: npx skills add opendndapps/ogt-skills/opendndapps-ogt-skills-ogt-docs-init

OGT Docs - Init

Initialize the docs-first folder structure for a new or existing project.

Overview

This skill creates the complete docs/ folder hierarchy for a docs-first project. It sets up folders for definitions, tasks, rules, and all supporting structure.

flowchart TB
    INIT["ogt docs init"] --> DOCS["docs/"]

    DOCS --> DEF["define/"]
    DOCS --> TODO["todo/"]
    DOCS --> RULES["rules/"]
    DOCS --> GUIDES["guides/"]

    DEF --> |business, features, code...| DEF_SUB[...]
    TODO --> |pending, in_progress, done...| TODO_SUB[...]
    RULES --> |code, git, docs...| RULES_SUB[...]

When to Use

  • Setting up a brand new project
  • Adding docs-first structure to existing project
  • Migrating from another documentation system
  • Onboarding team to docs-first workflow

Complete Folder Structure

docs/
├── define/                         # Definitions (WHAT)
│   ├── business/                   # Business model
│   │   ├── pricing/
│   │   ├── users/
│   │   ├── revenue/
│   │   └── positioning/
│   │
│   ├── features/                   # Product features
│   │   └── {feature_name}/
│   │       ├── feature.md
│   │       ├── mvp.md
│   │       ├── phase_0.md
│   │       └── ...
│   │
│   ├── code/                       # Technical definitions
│   │   ├── architecture/
│   │   ├── services/
│   │   ├── data_models/
│   │   └── api/
│   │
│   ├── marketing/                  # Marketing definitions
│   │   ├── value_proposition/
│   │   ├── personas/
│   │   ├── messaging/
│   │   └── positioning/
│   │
│   ├── branding/                   # Brand identity
│   │   ├── visual_identity/
│   │   ├── voice_tone/
│   │   └── guidelines/
│   │
│   └── tools/                      # Tool documentation
│       ├── cli/
│       ├── scripts/
│       └── third_party/
│
├── todo/                           # Tasks (DOING)
│   ├── pending/                    # Not started
│   ├── in_progress/                # Being worked on
│   ├── review/                     # Awaiting review
│   ├── blocked/                    # Cannot proceed
│   ├── done/                       # Completed
│   ├── rejected/                   # Declined
│   ├── implemented/                # Deployed
│   └── audit_log/                  # Audit history
│
├── rules/                          # Standards (HOW)
│   ├── code/                       # Coding standards
│   │   ├── general/
│   │   ├── typescript/
│   │   ├── naming/
│   │   ├── front/
│   │   └── back/
│   │
│   ├── git/                        # Git workflow
│   │   ├── commits/
│   │   ├── branches/
│   │   ├── pull_requests/
│   │   └── reviews/
│   │
│   ├── docs/                       # Documentation standards
│   │   ├── structure/
│   │   └── formatting/
│   │
│   └── style/                      # Style guides
│       ├── ui/
│       └── api/
│
├── guides/                         # How-to guides
│   ├── getting_started/
│   ├── development/
│   └── deployment/
│
├── agents/                         # AI agent definitions
│   ├── _shared/                    # Shared context
│   │   ├── CONTEXT.md
│   │   ├── SKILLS.md
│   │   └── DOCS_STRUCTURE.md
│   │
│   └── {agent_name}/
│       └── AGENT.md
│
├── skills/                         # AI skills
│   └── {skill_name}/
│       └── SKILL.md
│
└── _templates/                     # Templates for new docs
    ├── task.md
    ├── feature.md
    ├── definition.md
    ├── rule.md
    └── tool.md

Initialization Steps

Step 1: Create Root Structure

mkdir -p docs/{define,todo,rules,guides,agents,skills,_templates}

Step 2: Create Define Subfolders

# Business
mkdir -p docs/define/business/{pricing,users,revenue,positioning}

# Features
mkdir -p docs/define/features

# Code
mkdir -p docs/define/code/{architecture,services,data_models,api}

# Marketing
mkdir -p docs/define/marketing/{value_proposition,personas,messaging,positioning}

# Branding
mkdir -p docs/define/branding/{visual_identity,voice_tone,guidelines}

# Tools
mkdir -p docs/define/tools/{cli,scripts,dev,third_party}

Step 3: Create Todo Subfolders

mkdir -p docs/todo/{pending,in_progress,review,blocked,done,rejected,implemented,audit_log}

Step 4: Create Rules Subfolders

# Code rules
mkdir -p docs/rules/code/{general,typescript,naming,errors,async,front,back}

# Git rules
mkdir -p docs/rules/git/{commits,branches,pull_requests,reviews,merging,tags}

# Docs rules
mkdir -p docs/rules/docs/{structure,formatting}

# Style rules
mkdir -p docs/rules/style/{ui,api}

Step 5: Create Guides Subfolders

mkdir -p docs/guides/{getting_started,development,deployment}

Step 6: Create Agent Structure

mkdir -p docs/agents/_shared

Step 7: Create Templates

See Templates section below.


Templates

docs/_templates/task.md

# Task: {Title}

## Summary

One paragraph describing what needs to be done and why.

## Objectives

- Specific objective 1
- Specific objective 2
- Specific objective 3

## Acceptance Criteria

- [ ] Verifiable criterion 1
- [ ] Verifiable criterion 2
- [ ] Verifiable criterion 3
- [ ] TypeScript compiles clean (if applicable)

## Dependencies

- {dependency} or "None"

## Estimated Effort

{Tiny|Small|Medium|Large|XLarge} ({time estimate})

## References

- Relevant link or file path

docs/_templates/feature.md

# Feature: {Name}

## Summary

One paragraph explaining the feature and its value.

## User Stories

As a {user type}, I want to {action}, so that {benefit}.

## Scope

### In Scope

- Item 1
- Item 2

### Out of Scope

- Item 1
- Item 2

## Success Metrics

- Metric 1
- Metric 2

## Phases

See mvp.md and phase files for implementation details.

docs/_templates/definition.md

# {Name}

## Summary

One paragraph explaining what this is.

## Details

Detailed explanation of the concept, entity, or component.

## Structure

{structure diagram or tree}


## Examples

### Example 1

{description and example}

### Example 2

{description and example}

## Related

- Link to related definition
- Link to related rule

docs/_templates/rule.md

# Rule: {Name}

## Summary

One sentence stating the rule clearly.

## Rationale

Why this rule exists.

## The Rule

Clear statement using RFC 2119 keywords (MUST, SHOULD, MAY).

## Examples

### Correct

{correct example}

### Incorrect

{incorrect example with explanation}

## Exceptions

Documented exceptions.

## Enforcement

How the rule is enforced.

docs/_templates/tool.md

# Tool: {Name}

## Summary

Brief description of the tool.

## Purpose

What problem it solves.

## Installation

```bash
{installation command}
```

Quick Start

{basic usage}

Commands

CommandDescription
{cmd1}{description}
{cmd2}{description}

Configuration

{configuration options}

Examples

{usage examples}


---

## Shared Agent Context

### docs/agents/_shared/CONTEXT.md

```markdown
# Project Context

## Project Name

{Project Name}

## Description

{One paragraph description}

## Tech Stack

- Frontend: {stack}
- Backend: {stack}
- Database: {database}
- Infrastructure: {infra}

## Repository

{repo URL}

## Key Directories

{project}/ ├── {dir1}/ # {purpose} ├── {dir2}/ # {purpose} └── docs/ # Documentation (docs-first)


## Current Focus

{What the team is currently working on}

## Key Decisions

1. {Decision 1} - {rationale}
2. {Decision 2} - {rationale}

docs/agents/_shared/SKILLS.md

# Available Skills

Skills available to all agents in this project.

## Documentation Skills

| Skill                | Trigger         | Purpose            |
| -------------------- | --------------- | ------------------ |
| ogt-docs-create-task | "create a task" | Create new task    |
| ogt-docs-define      | "define X"      | Create definitions |
| ogt-docs-rules       | "create rule"   | Create rules       |
| ogt-docs-audit-task  | "audit tasks"   | Verify tasks       |

## Code Skills

{project-specific code skills}

## Tool Skills

{project-specific tool skills}

docs/agents/_shared/DOCS_STRUCTURE.md

# Documentation Structure

## Folder Convention

Every documentable item is a **folder**, not a file.

docs/{section}/{category}/{item_name}/ ├── {type}.md # Primary document (task.md, feature.md, etc.) ├── {supporting}.md # Supporting documents └── .{signal} # Signal files


## Signal Files

| Signal | Type | Purpose |
|--------|------|---------|
| `.version` | Content | Schema version |
| `.priority` | Content | Priority level |
| `.blocked` | Empty | Item is blocked |
| `.verified` | Empty | Item verified |

## Workflow

Tasks move between folders:

pending/ -> in_progress/ -> review/ -> done/ -> implemented/


Features evolve through phases:

mvp.md -> phase_0.md -> phase_1.md -> ...


Init Script

Complete initialization script:

#!/bin/bash
# init-docs.sh - Initialize docs-first structure

set -e

echo "Initializing docs-first structure..."

# Create main directories
mkdir -p docs/{define,todo,rules,guides,agents,skills,_templates}

# Define subdirectories
mkdir -p docs/define/business/{pricing,users,revenue,positioning}
mkdir -p docs/define/features
mkdir -p docs/define/code/{architecture,services,data_models,api}
mkdir -p docs/define/marketing/{value_proposition,personas,messaging,positioning}
mkdir -p docs/define/branding/{visual_identity,voice_tone,guidelines}
mkdir -p docs/define/tools/{cli,scripts,dev,third_party}

# Todo subdirectories
mkdir -p docs/todo/{pending,in_progress,review,blocked,done,rejected,implemented,audit_log}

# Rules subdirectories
mkdir -p docs/rules/code/{general,typescript,naming,errors,async,front,back}
mkdir -p docs/rules/git/{commits,branches,pull_requests,reviews,merging,tags}
mkdir -p docs/rules/docs/{structure,formatting}
mkdir -p docs/rules/style/{ui,api}

# Guides subdirectories
mkdir -p docs/guides/{getting_started,development,deployment}

# Agent structure
mkdir -p docs/agents/_shared

# Create template files
cat > docs/_templates/task.md << 'EOF'
# Task: {Title}

## Summary

One paragraph describing what needs to be done and why.

## Objectives

- Specific objective 1
- Specific objective 2

## Acceptance Criteria

- [ ] Verifiable criterion 1
- [ ] Verifiable criterion 2
- [ ] TypeScript compiles clean

## Dependencies

None

## Estimated Effort

Medium (2-4 hours)
EOF

cat > docs/_templates/feature.md << 'EOF'
# Feature: {Name}

## Summary

One paragraph explaining the feature.

## User Stories

As a user, I want to {action}, so that {benefit}.

## Scope

### In Scope

- Item 1

### Out of Scope

- Item 1

## Success Metrics

- Metric 1
EOF

cat > docs/_templates/definition.md << 'EOF'
# {Name}

## Summary

One paragraph explaining what this is.

## Details

Detailed explanation.

## Examples

Example 1

## Related

- Related link
EOF

cat > docs/_templates/rule.md << 'EOF'
# Rule: {Name}

## Summary

One sentence stating the rule.

## Rationale

Why this rule exists.

## The Rule

MUST/SHOULD/MAY statement.

## Examples

### Correct

Example

### Incorrect

Example

## Enforcement

How enforced.
EOF

# Create shared agent context
cat > docs/agents/_shared/CONTEXT.md << 'EOF'
# Project Context

## Project Name

{Project Name}

## Description

{Description}

## Tech Stack

- Frontend:
- Backend:
- Database:

## Key Directories

project/ ├── src/ └── docs/

EOF

cat > docs/agents/_shared/DOCS_STRUCTURE.md << 'EOF'
# Documentation Structure

Every documentable item is a folder, not a file.

## Signal Files

| Signal | Purpose |
|--------|---------|
| `.version` | Schema version |
| `.priority` | Priority level |
| `.verified` | Item verified |
EOF

echo "Done! Structure created in docs/"
echo ""
echo "Next steps:"
echo "1. Edit docs/agents/_shared/CONTEXT.md with project info"
echo "2. Create initial feature definitions in docs/define/features/"
echo "3. Add first tasks to docs/todo/pending/"

Verification

After initialization, verify the structure:

# Count directories created
find docs -type d | wc -l
# Expected: ~60 directories

# Check key directories exist
test -d docs/define/features && echo "features: OK"
test -d docs/todo/pending && echo "pending: OK"
test -d docs/rules/code && echo "code rules: OK"
test -d docs/_templates && echo "templates: OK"

# Check templates exist
test -f docs/_templates/task.md && echo "task template: OK"
test -f docs/_templates/feature.md && echo "feature template: OK"

Post-Init Steps

1. Configure Project Context

Edit docs/agents/_shared/CONTEXT.md:

vim docs/agents/_shared/CONTEXT.md

Fill in:

  • Project name
  • Description
  • Tech stack
  • Key directories

2. Create First Feature

mkdir -p docs/define/features/user_auth
cp docs/_templates/feature.md docs/define/features/user_auth/feature.md
vim docs/define/features/user_auth/feature.md

3. Create First Task

mkdir -p docs/todo/pending/setup_project
cp docs/_templates/task.md docs/todo/pending/setup_project/task.md
vim docs/todo/pending/setup_project/task.md

4. Add Essential Rules

Start with commit message format:

mkdir -p docs/rules/git/commits
cp docs/_templates/rule.md docs/rules/git/commits/rule.md
vim docs/rules/git/commits/rule.md

Customization

Minimal Init

For smaller projects, use minimal structure:

mkdir -p docs/{define/features,todo/{pending,done},rules/code,_templates}

Extended Init

For larger projects, add:

# Multiple environments
mkdir -p docs/define/environments/{dev,staging,prod}

# Multiple teams
mkdir -p docs/todo/{frontend,backend,devops}

# Detailed rules
mkdir -p docs/rules/code/{react,node,database}

Signal Files Reference

SignalCreated DuringContent
.versionFirst document{"schema": "1.0"}
.initializedInitTimestamp
.last_auditAuditTimestamp

Init Checklist

  • All main directories created
  • Define subdirectories created
  • Todo workflow directories created
  • Rules categories created
  • Templates created
  • Shared agent context created
  • CONTEXT.md filled in
  • First feature defined
  • First task created
  • Essential rules documented

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Automation

ogt-agent-team

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

ogt-agent-spawn

No summary provided by upstream source.

Repository SourceNeeds Review
Web3

ogt-docs-define-tools

No summary provided by upstream source.

Repository SourceNeeds Review
Coding

ogt-cli-claude

No summary provided by upstream source.

Repository SourceNeeds Review