Skip to content

benthongtiang/AgentDeploy

Repository files navigation

AgentDeploy (ADLC Auto-Deploy Pipeline)

An agentic DevOps & Auto-Deployment Pipeline toolset (operating via the .adlc/ directory inside consumer projects) for automated codebase analysis, environment provisioning, deployment execution, and self-healing system recovery on cloud providers or VPS environments.


Table of Contents

  1. Pipeline Overview
  2. Directory Structure
  3. Global Setup & Short Commands
  4. Phase Details
  5. Core Ethos

Pipeline Overview

When executing the /deploy command, the system triggers sub-skills sequentially as follows:

graph TD
    A[Start /deploy] --> B[Phase 0: Preflight & Resolve Target]
    B --> C[Phase 1: Codebase Analysis /deploy-analyze]
    C --> D[Phase 2: Environment Config /deploy-env]
    D --> E[Phase 3: Service Provisioning /deploy-provision]
    E --> F[Phase 4: Trigger & Build Monitor /deploy-trigger]
    F -->|Success| G[Phase 6: Health Verification & Handover]
    F -->|Failure| H[Phase 5: Self-Healing /deploy-heal]
    H -->|Fix Applied| F
    H -->|Retry Exceeded 3 times| I[Halt & Report Error]
Loading

Directory Structure

AgentDeploy/
├── README.md               # Toolset documentation
├── ETHOS.md                # Agent rules and core principles
├── deploy/                 # Main orchestrator /deploy
│   └── SKILL.md
├── deploy-analyze/         # Codebase analysis & config generator
│   └── SKILL.md
├── deploy-env/             # Environment variable detection & setup helper
│   └── SKILL.md
├── deploy-provision/       # Service & database provisioning on Cloud/VPS
│   └── SKILL.md
├── deploy-trigger/         # Build execution & log analyzer
│   └── SKILL.md
├── deploy-heal/            # Failure analysis & self-healing patcher
│   └── SKILL.md
└── partials/               # Shared scripts across phases
    └── ethos-include.sh

Global Setup & Short Commands

To execute short commands such as /deploy or /deploy-heal in your IDE/AI Agent and have the development tools automatically load this toolset's sequence, follow the configuration instructions for your preferred AI tool below:

🌌 1. Google Gemini / Custom Agent Rules

  • File Location: ~/.gemini/GEMINI.md or locally in .agent/rules/ within your project folder.
  • Configuration (Paths mapped to this directory /path/to/AgentDeploy):
When the user inputs a short command, always execute the `view_file` tool in the background with `IsSkillFile: true` pointing to the corresponding Skill file before starting work:
- `/deploy` -> Read the file `/path/to/AgentDeploy/deploy/SKILL.md`
- `/deploy-analyze` -> Read the file `/path/to/AgentDeploy/deploy-analyze/SKILL.md`
- `/deploy-env` -> Read the file `/path/to/AgentDeploy/deploy-env/SKILL.md`
- `/deploy-provision` -> Read the file `/path/to/AgentDeploy/deploy-provision/SKILL.md`
- `/deploy-trigger` -> Read the file `/path/to/AgentDeploy/deploy-trigger/SKILL.md`
- `/deploy-heal` -> Read the file `/path/to/AgentDeploy/deploy-heal/SKILL.md`

🚀 2. Cursor

  • File Location: Create a .cursorrules file at the root of your project.
  • Configuration:
When the user inputs a short command, always read the corresponding Skill file immediately before taking action or responding:
- `/deploy` -> Read the file `/path/to/AgentDeploy/deploy/SKILL.md`
- `/deploy-analyze` -> Read the file `/path/to/AgentDeploy/deploy-analyze/SKILL.md`
- `/deploy-env` -> Read the file `/path/to/AgentDeploy/deploy-env/SKILL.md`
- `/deploy-provision` -> Read the file `/path/to/AgentDeploy/deploy-provision/SKILL.md`
- `/deploy-trigger` -> Read the file `/path/to/AgentDeploy/deploy-trigger/SKILL.md`
- `/deploy-heal` -> Read the file `/path/to/AgentDeploy/deploy-heal/SKILL.md`

🏄 3. Windsurf

  • File Location: Create a .windsurfrules file at the root of your project.
  • Configuration:
When the user inputs a short command or references these deployment processes, read the corresponding Skill file immediately before proceeding:
- `/deploy` -> Read the file `/path/to/AgentDeploy/deploy/SKILL.md`
- `/deploy-analyze` -> Read the file `/path/to/AgentDeploy/deploy-analyze/SKILL.md`
- `/deploy-env` -> Read the file `/path/to/AgentDeploy/deploy-env/SKILL.md`
- `/deploy-provision` -> Read the file `/path/to/AgentDeploy/deploy-provision/SKILL.md`
- `/deploy-trigger` -> Read the file `/path/to/AgentDeploy/deploy-trigger/SKILL.md`
- `/deploy-heal` -> Read the file `/path/to/AgentDeploy/deploy-heal/SKILL.md`

🛠️ 4. Roo Code / Cline (VS Code Extension)

  • File Location: Create a .clinerules file at the root of your project.
  • Configuration:
When the user inputs a short command, read the corresponding Skill file before starting work:
- `/deploy` -> Read the file `/path/to/AgentDeploy/deploy/SKILL.md`
- `/deploy-analyze` -> Read the file `/path/to/AgentDeploy/deploy-analyze/SKILL.md`
- `/deploy-env` -> Read the file `/path/to/AgentDeploy/deploy-env/SKILL.md`
- `/deploy-provision` -> Read the file `/path/to/AgentDeploy/deploy-provision/SKILL.md`
- `/deploy-trigger` -> Read the file `/path/to/AgentDeploy/deploy-trigger/SKILL.md`
- `/deploy-heal` -> Read the file `/path/to/AgentDeploy/deploy-heal/SKILL.md`

🐙 5. GitHub Copilot

  • File Location: Create a .github/copilot-instructions.md file at the root of your project.
  • Configuration:
When the user references these custom slash commands, read the content of the referenced file to understand the skill context:
- `/deploy` -> Read the file `/path/to/AgentDeploy/deploy/SKILL.md`
- `/deploy-analyze` -> Read the file `/path/to/AgentDeploy/deploy-analyze/SKILL.md`
- `/deploy-env` -> Read the file `/path/to/AgentDeploy/deploy-env/SKILL.md`
- `/deploy-provision` -> Read the file `/path/to/AgentDeploy/deploy-provision/SKILL.md`
- `/deploy-trigger` -> Read the file `/path/to/AgentDeploy/deploy-trigger/SKILL.md`
- `/deploy-heal` -> Read the file `/path/to/AgentDeploy/deploy-heal/SKILL.md`

💎 6. JetBrains AI Assistant (IntelliJ, WebStorm, PyCharm, etc.)

  • File Location: Add the rules to your custom System Prompt / Prompt Library, or create a .github/copilot-instructions.md file at the root of your project.
  • Configuration:
When the user inputs a short command, refer to the corresponding Skill file:
- `/deploy` -> Read the file `/path/to/AgentDeploy/deploy/SKILL.md`
- `/deploy-analyze` -> Read the file `/path/to/AgentDeploy/deploy-analyze/SKILL.md`
- `/deploy-env` -> Read the file `/path/to/AgentDeploy/deploy-env/SKILL.md`
- `/deploy-provision` -> Read the file `/path/to/AgentDeploy/deploy-provision/SKILL.md`
- `/deploy-trigger` -> Read the file `/path/to/AgentDeploy/deploy-trigger/SKILL.md`
- `/deploy-heal` -> Read the file `/path/to/AgentDeploy/deploy-heal/SKILL.md`

🤖 7. Claude Code

  • File Location: Create a .clauderules file at the root of your project.
  • Configuration:
When the user inputs a short command, always read the corresponding Skill file immediately before taking action or responding:
- `/deploy` -> Read the file `/path/to/AgentDeploy/deploy/SKILL.md`
- `/deploy-analyze` -> Read the file `/path/to/AgentDeploy/deploy-analyze/SKILL.md`
- `/deploy-env` -> Read the file `/path/to/AgentDeploy/deploy-env/SKILL.md`
- `/deploy-provision` -> Read the file `/path/to/AgentDeploy/deploy-provision/SKILL.md`
- `/deploy-trigger` -> Read the file `/path/to/AgentDeploy/deploy-trigger/SKILL.md`
- `/deploy-heal` -> Read the file `/path/to/AgentDeploy/deploy-heal/SKILL.md`

Phase Details

🚀 /deploy (Orchestrator)

Coordinates the execution of all sub-skills, starting from credential checks to final deployment verification.

🔍 /deploy-analyze

  • Analyzes the active tech stack (Node.js, Python, Go, Rust, Ruby, etc.).
  • Detects network ports used by the application and required databases.
  • Automatically generates a Dockerfile, docker-compose.yml, or nixpacks.toml if they do not already exist.

🔑 /deploy-env

  • Locates .env.example files or scans source code for environment variables.
  • Summarizes the required environment variables and prompts the user to enter secrets/default configurations securely.

🛠️ /deploy-provision

  • Connects to deployment engines (Coolify API, Railway CLI) or VPS (via SSH).
  • Provisions databases (PostgreSQL, Redis, MySQL, etc.) as identified in the analysis phase.
  • Binds domains, configures port forwarding, and attaches environment variables to the app service.

🎬 /deploy-trigger

  • Triggers the build and deployment process on Cloud/VPS.
  • Streams build logs and server outputs in real-time.
  • Performs a health check by sending HTTP GET requests to the deployment URL.

🏥 /deploy-heal (Self-Healing)

  • If the deployment fails, the agent fetches and analyzes the error logs.
  • Instantly fixes root causes in source code or configuration files.
  • Commits, pushes changes, and triggers a redeployment automatically (limits to 3 attempts to prevent infinite loops).

Core Ethos

For developer guidelines and standard operating procedures, please refer to ETHOS.md.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages