Skip to content

README.md

Latest commit

 

History

History
128 lines (91 loc) · 4.88 KB

README.md

File metadata and controls

128 lines (91 loc) · 4.88 KB

Project Documenter Plugin

Generate professional project documentation with draw.io architecture diagrams and Word (.docx) output with embedded PNG images. Works on any software project — automatically discovers the technology stack, architecture, and code structure.

Installation

# Using Copilot CLI
copilot plugin install project-documenter@awesome-copilot

What It Does

Point the Project Documenter agent at any repository and it produces:

  1. Markdown document — 10-section project summary with embedded diagram references
  2. Draw.io diagrams — C4 Context, Pipeline, and Component relationship diagrams (.drawio + .drawio.png)
  3. Word document — professionally formatted .docx with title page, table of contents, and embedded PNG architecture images

Agent Workflow

What's Included

Agent

Agent Description
project-documenter Generates professional project documentation with draw.io architecture diagrams and Word document output with embedded images. Auto-discovers any project's technology stack and architecture.

Skills

Skill Description
drawio Generate draw.io diagrams as .drawio files and export to PNG via bundled Node.js script (uses draw.io CLI or headless browser)
md-to-docx Convert Markdown to Word (.docx) with embedded PNG images — pure JavaScript, no Pandoc required

How It Works

Step 1: Discover

The agent scans your repository to understand:

  • Technology stack (.csproj, package.json, pom.xml, go.mod, etc.)
  • Architecture pattern (API, worker service, CLI, library)
  • Design patterns (factory, strategy, repository, pipeline)
  • Interfaces, implementations, models, configuration
  • Dependencies, Docker setup, CI/CD

Step 2: Generate Diagrams

Creates 3-5 professional draw.io diagrams following the C4 Model:

Diagram C4 Level Shows
High-Level Architecture Context System in its environment — upstream, downstream, external deps
Processing Pipeline Container Internal data flow — entry point → stages → output
Component Relationships Component Interfaces, implementations, factories, DI graph
Deployment (optional) Infrastructure Docker, Kubernetes, scaling, cloud services
Data Model (optional) Component Entity/DTO hierarchy (if significant)

Each diagram is exported to PNG using the bundled drawio-to-png.mjs script.

Step 3: Write Markdown

Produces docs/project-summary.md with 10 sections:

  1. Executive Summary
  2. Architecture Overview (with embedded diagram)
  3. Processing Pipeline (with embedded diagram)
  4. Core Components (with embedded diagram)
  5. API Contracts / Message Schemas
  6. Infrastructure & Deployment
  7. Extension Patterns
  8. Rules & Anti-Patterns
  9. Dependencies
  10. Code Structure

Step 4: Word Document

Converts the Markdown to a formatted .docx using the bundled md-to-docx.mjs script:

  • Title page with project name, date, version, audience
  • Auto-generated table of contents
  • PNG diagram images embedded inline in the Word document
  • Calibri font, colored headings, styled tables with alternating rows
  • Code blocks in Consolas with shaded background

Step 5: Verify

Spot-checks class names, file paths, and diagram accuracy against the actual codebase. Reports all generated files.

Generated Output

docs/
├── project-summary.md                     # Source document (Markdown)
├── project-summary.docx                   # Word document with embedded images
└── diagrams/
    ├── high-level-architecture.drawio     # C4 Context diagram (editable)
    ├── high-level-architecture.drawio.png # Rendered PNG
    ├── processing-pipeline.drawio         # C4 Container diagram
    ├── processing-pipeline.drawio.png
    ├── component-relationships.drawio     # C4 Component diagram
    └── component-relationships.drawio.png

Prerequisites

Requirement Purpose Required?
Node.js 18+ Run bundled export scripts Yes
Edge or Chrome Headless browser for diagram rendering One of: this OR draw.io desktop
draw.io desktop CLI diagram export (faster alternative) Optional (browser fallback available)

Technology Agnostic

Works with any stack. The agent auto-detects:

  • .NET (.csproj, .sln), Java (pom.xml, build.gradle), Node.js (package.json), Python (pyproject.toml), Go (go.mod), Rust (Cargo.toml)
  • Docker, Kubernetes, GitHub Actions, GitLab CI
  • Any messaging system (SQS, RabbitMQ, Kafka, Azure Service Bus)
  • Any database ORM (EF, Hibernate, Prisma, SQLAlchemy)

Source

This plugin is part of Awesome Copilot, a community-driven collection of GitHub Copilot extensions.

License

MIT