with one click
with one click
| type | skill |
| name | Documentation |
| description | Generate and update technical documentation |
| skillSlug | documentation |
| phases | ["P","C"] |
| generated | "2026-03-18T00:00:00.000Z" |
| status | filled |
| scaffoldVersion | 2.0.0 |
Guidelines for creating and maintaining documentation in the @dotcontext/cli project.
.context/plans/.context/docs/.context/ content (docs, agents, skills, plans)| Location | Purpose | Format |
|---|---|---|
.context/docs/ | Project knowledge base (generated scaffolds) | Markdown with YAML frontmatter |
.context/agents/ | AI agent playbooks | Markdown with YAML frontmatter |
.context/skills/ | On-demand expertise guides | Markdown with YAML frontmatter |
.context/plans/ | Feature plans and tracking | Markdown with YAML frontmatter |
README.md | Public-facing project overview | Standard Markdown |
CHANGELOG.md | Version history (Keep a Changelog format) | Standard Markdown |
CONTRIBUTING.md | Contributor guidelines | Standard Markdown |
AGENTS.md | Agent knowledge base reference | Standard Markdown |
CLAUDE.md | Claude-specific project instructions | Standard Markdown |
Scaffold docs use v2 frontmatter:
---
type: documentation
name: Project Overview
description: High-level project summary
category: core
generated: 2026-03-18
status: filled
scaffoldVersion: "2.0.0"
---
Core docs in this project:
project-overview.md -- Technology stack, entry points, getting starteddevelopment-workflow.md -- Branching, CI, contributing processtesting-strategy.md -- Jest config, test types, quality gatestooling.md -- CLI scripts, dev commands, IDE setupcodebase-map.json -- Auto-generated semantic analysis outputWhen updating, preserve the frontmatter and set status: filled.
Each playbook follows this structure:
Available agents: code-reviewer, bug-fixer, feature-developer, refactoring-specialist, test-writer, documentation-writer, performance-optimizer.
Skills use this structure:
Follow Keep a Changelog format. Group changes under version headers:
## [0.7.2]
### Included Pull Requests
- [#XX](url) - type: description
### Added
- **Feature Name**: Details
### Changed
- **Area**: What changed
### Fixed
- **Bug**: What was fixed
Plans are created via context({ action: 'scaffoldPlan', planName: '...' }) or manually. Structure:
---
type: plan
name: feature-name
planSlug: feature-name
status: active
---
# Feature Name
## Goal
What this achieves.
## Phases
### Phase 1: Planning
- [ ] Step 1
- [ ] Step 2
### Phase 2: Execution
- [ ] Step 1
src/services/mcp/mcpServer.ts, not "the MCP server file"npx tsx src/index.ts <command>--- delimiterssrc/utils/i18n.tsen and pt-BR localesStandalone CLI AI generation is not supported. Documentation completion flows through MCP:
context({ action: 'fillSingle', filePath: '/path/to/doc.md' }) -- returns semantic context and scaffold structure for one filecontext({ action: 'fill', target: 'docs' }) -- returns batched guidance for multiple scaffold filessrc/services/autoFill/autoFillService.ts -- provides static starter content during scaffold generationThe MCP fill process:
context({ action: 'init' })SemanticContextBuilderstatus: filledDesign MCP tools and gateway interfaces for the dotcontext server
Systematic bug investigation and root cause analysis
Review code quality, patterns, and best practices
Generate commit messages following conventional commits with scope detection
Break down features into implementable tasks
Review pull requests against team standards and best practices