diagram-to-iac 1.19.0

Convert architecture diagrams into IaC modules

diagram-to-iac

"One container, many minds—zero manual toil."

An automated DevOps-in-a-Box system that intelligently handles complete Repo-to-Deployment (R2D) workflows. The project combines AI-powered infrastructure analysis with GitHub automation for self-healing deployments.

🚀 DevOps-in-a-Box: R2D Action

The R2D (Repo-to-Deployment) Action is a self-healing, Terraform-first DevOps automation solution powered by a sophisticated multi-agent system. Integrated via a GitHub Actions reusable workflow, it analyzes your repository and deploys your infrastructure, primarily using Terraform Cloud.

🎯 Quick Start

Integrate the R2D Action into your repository in minutes:

Step 1: Add the Reusable Workflow File

Create .github/workflows/devops-in-a-box.yml (or your preferred name) in your repository:

name: "🤖 DevOps-in-a-Box: R2D Deployment"

on:
  workflow_dispatch: # Allows manual triggering
    inputs:
      repo_url:
        description: 'Repository URL to deploy (defaults to current repo)'
        required: false
        default: ''
        type: string
      branch:
        description: 'Branch to deploy (defaults to main/master)'
        required: false
        default: ''
        type: string
      dry_run:
        description: 'Run in dry-run mode (true/false)'
        required: false
        default: 'true'
        type: boolean
      trigger_label:
        description: 'Issue label for issue triggers'
        required: false
        default: 'r2d-request'
        type: string
  issues:
    types: [labeled] # Trigger on issue labeling
  pull_request:
    types: [closed] # Trigger on PR merge

jobs:
  r2d_deployment:
    name: "🚀 R2D Deployment"
    uses: "amartyamandal/diagram-to-iac/.github/workflows/r2d-unified.yml@main"
    secrets: inherit # Passes your repository secrets to the reusable workflow
    with:
      repo_url: ${{ inputs.repo_url || format('{0}/{1}', github.server_url, github.repository) }}
      branch: ${{ inputs.branch }}
      dry_run: ${{ inputs.dry_run }}
      trigger_label: ${{ inputs.trigger_label }}

Step 2: Configure Required Secrets

Go to your repository's Settings → Secrets and variables → Actions and add the necessary secrets. All secrets must be base64-encoded.

Key Secrets:

• REPO_API_KEY: Your GitHub Personal Access Token (PAT) for repository access and GHCR authentication.
• TF_API_KEY: Your Terraform Cloud API Token.
• An AI Provider API Key (e.g., OPENAI_API_KEY).

📚 For detailed instructions on secret setup (including PAT scopes and base64 encoding) and more, see the Complete R2D User Guide.

Step 3: Trigger Deployment You can now deploy via manual dispatch, by labeling a GitHub issue (e.g., with r2d-request), or by merging a Pull Request.

Key Features

• 🤖 AI-Powered & Self-Healing: Automatically analyzes repositories, creates GitHub Issues for errors, and can suggest fixes.
• 🛡️ Secure & Isolated: Runs in a non-root container with workspace isolation and command allowlisting.
• 🌍 IaC Focused: Primarily supports Terraform deployments via Terraform Cloud, with capabilities for Ansible, PowerShell, and Bash.
• 📊 Observable: Provides rich logging, step summaries, and deployment artifacts.
• 🔄 Flexible Triggers: Supports manual, issue-based, and PR-merge triggers.

The Cast: Specialized Agents

Our multi-agent system efficiently handles different aspects of the DevOps lifecycle:

Agent	Capability	Never Does
SupervisorAgent	Orchestrates workflow, manages state, routes tasks	Edit code directly
GitAgent	Clones repos, manages branches, creates PRs	Guess network credentials
ShellAgent	Executes commands safely, detects stack	Run non-allowlisted binaries
TerraformAgent	init/plan/apply, classifies errors, cost scans	Apply with critical sec issues
PolicyAgent	Runs tfsec + OPA for security gates	Ignore critical findings
VisionAgent	(If applicable) Analyzes diagrams for IaC generation	-

(For more details on agents, see docs/AGENT_ARCHITECTURE.md)

📦 Installation (for CLI usage or development)

If you want to run the diagram-to-iac CLI tools locally or contribute to development:

# Clone the repository
git clone https://github.com/amartyamandal/diagram-to-iac.git
cd diagram-to-iac

# Create a virtual environment (Python 3.11+) and install
python -m venv .venv
source .venv/bin/activate
pip install -e ".[dev]"

For detailed developer setup, see the Developer Guide.

🖥️ Running the CLI (for local use)

The project provides CLI entry points for direct interaction with agents (primarily for development and testing):

# Main R2D CLI (simulates parts of the GitHub Action flow)
diagram-to-iac --repo-url https://github.com/user/repo --dry-run

# Get help
diagram-to-iac --help

# Individual agent CLIs (examples)
supervisor-agent --repo-url https://github.com/user/repo
git-agent --repo-url https://github.com/user/repo --branch my-feature

📊 Observability

R2D ensures you're informed about its operations:

• GitHub Issue Updates: Detailed progress and results are posted in the triggering GitHub Issue.
• Structured Logging: JSONL logs are generated for each run (available as workflow artifacts).
• Step Summary Dashboard: A step-summary.md file is generated with Terraform changes, security findings, and metrics (available as a workflow artifact).
• GitHub Actions Artifacts: Logs, plans, and reports are uploaded for download from the workflow run page.

📚 Documentation

• 🎯 R2D User Guide: Your primary guide for setting up and using the R2D Action.
• 🛠️ Developer Guide: For contributors or those wanting to understand the internals and run tools locally.
• 🤖 Agent Architecture: Learn about the different AI agents and how they work.
• 💡 DevOps-in-a-Box: The Severance Protocol for Infrastructure: A conceptual overview of the project's philosophy.

🤝 Contributing

Contributions are welcome! Please see our development guidelines (in docs/DEVELOPER_GUIDE.md and .github/copilot-instructions.md) for coding standards and contribution instructions.

---

"One container, many minds—zero manual toil." 🤖