Digital FTE

🤖 Digital-FTE — Abdullah Junior

> "Your life and business on autopilot. A self-evolving, cloud-native AI Employee that never sleeps."

[](https://github.com/AbdullahMalik17/Digital-FTE/actions/workflows/python-ci.yml) [](https://github.com/AbdullahMalik17/Digital-FTE) [](https://github.com/AbdullahMalik17/Digital-FTE) [](https://opensource.org/licenses/MIT) [](https://github.com/AbdullahMalik17/Digital-FTE/stargazers) [](https://github.com/AbdullahMalik17/Digital-FTE/commits/main) [](https://github.com/AbdullahMalik17/Digital-FTE)

• Docs: Run npx mintlify dev docs (see docs/README.md) and add your deployed docs URL here.

---

🎬 What Is This?

Digital-FTE is an open-source autonomous AI agent system that acts as a 24/7 Digital Employee for your personal life and business. It watches your Gmail, WhatsApp, and LinkedIn — drafts replies, creates invoices, posts to social media, and executes tasks — all with human approval before any action is taken.

It uses a Dual-Agent Architecture (Cloud Sentry + Local Executive) connected by an Obsidian Vault as its central nervous system. The system can also debug and evolve itself — making it one of the first self-evolving AI employee implementations in open source.

📬 Email Arrives → 🤖 Cloud Agent Drafts Reply → 📂 Saved to Vault
→ 📱 You Approve on Phone → ⚡ Local Agent Executes → ✅ Done

---

✨ Key Features

• 📧 Gmail Watcher — Monitors inbox, categorizes emails, drafts context-aware replies
• 🏢 Odoo Accounting — Auto-creates invoices, logs expenses from email receipts
• 📱 Social Media Autopilot — Posts to Facebook, Instagram, Twitter/X with analytics
• 🧠 Smart Orchestrator — Routes tasks to the best AI model (Claude 3.5, Gemini 1.5 Pro, etc.)
• 🔐 Human-in-the-Loop — Nothing executes without your approval. Ever.
• 🔄 Self-Evolution Engine — Catches its own errors and writes its own patches
• 📂 Obsidian Vault Sync — Git-synced knowledge base as the system's memory
• 🐳 Docker Ready — One command deployment with docker-compose
• 🧪 Full Test Suite — Unit, integration, and E2E tests included

---

🏗️ Architecture

The system runs two agents in parallel:

☁️ Cloud Agent (The "Sentry")

• Lives: On any cloud server or always-on machine
• Watches: Gmail (IMAP), LinkedIn API, WhatsApp Web
• Does: Reads incoming messages, drafts responses using AI, writes proposals to the Vault
• Cannot: Access financial credentials or execute anything — read-only by design

💻 Local Agent (The "Executive")

• Lives: On your personal machine
• Syncs: Git-pulls the Vault every 60 seconds
• Does: Notifies you of pending tasks, waits for your approval, then executes via MCP servers
• Has access to: Odoo, social media APIs, local files, sensitive credentials

🧠 The Orchestrator

The brain that routes every task to the right tool and AI model based on complexity and cost.

┌─────────────────────────────────────────────────┐
│                  DIGITAL FTE LOOP               │
│                                                 │
│  📬 Input → ☁️ Cloud Agent → 📂 Vault (Git)     │
│                                  ↓              │
│            ✅ You Approve → 💻 Local Agent       │
│                                  ↓              │
│         ⚡ MCP Execution → 📁 Archive to Done   │
└─────────────────────────────────────────────────┘

---

🔌 Integrations (via MCP Servers)

| Service | What It Does | |---|---| | Gmail | Reads, categorizes, drafts context-aware replies | | Odoo | Creates invoices, logs vendor bills, financial summaries | | Facebook / Instagram | Posts content, fetches engagement analytics | | Twitter / X | Posts tweets/threads, monitors mentions | | WhatsApp | Reads messages, drafts replies | | Playwright | Browser automation for web tasks |

---

🚀 Quick Start

Prerequisites

• Python 3.10+
• Node.js 18+
• Docker (optional but recommended)
• An Obsidian vault (for the Vault sync)

1. Clone the repo

git clone https://github.com/AbdullahMalik17/Digital-FTE.git
cd Digital-FTE

2. Install dependencies

pip install -r requirements.txt

3. Configure environment

cp .env.example .env
# Fill in your API keys (see Configuration section below)

4. Start the system

# Start the local agent
python src/local_agent.py

# In a separate terminal, start the watchers
python src/service_manager.py

5. (Optional) Docker deployment

docker-compose up --build

---

⚙️ Configuration

Edit your .env file with the following:

# AI Models
OPENAI_API_KEY=your_key
ANTHROPIC_API_KEY=your_key
GOOGLE_API_KEY=your_key

# Gmail
GMAIL_CLIENT_ID=your_id
GMAIL_CLIENT_SECRET=your_secret

# Odoo (Accounting)
ODOO_URL=https://your-odoo-instance.com
ODOO_USERNAME=your_email
ODOO_PASSWORD=your_password

# Social Media
META_ACCESS_TOKEN=your_token
FACEBOOK_PAGE_ID=your_page_id
INSTAGRAM_ACCOUNT_ID=your_account_id
TWITTER_API_KEY=your_key
TWITTER_API_SECRET=your_secret
TWITTER_BEARER_TOKEN=your_token

---

📂 Vault Structure

Your Obsidian Vault is organized by task state:

| Folder | Purpose | |---|---| | Needs_Action/ | New incoming triggers awaiting AI draft | | Pending_Approval/ | Drafted tasks waiting for your approval | | In_Progress/ | Tasks being actively executed | | Done/ | Permanent archive of every completed action | | Logs/Audit/ | Full JSONL audit trail of every AI call | | Plans/ | Long-term strategic goals and breakdowns |

---

🧪 Testing

# Run full test suite
pytest tests/

# Individual test modules
pytest tests/test_odoo_integration.py      # Odoo connection + invoices
pytest tests/test_social_media_apis.py    # Meta + Twitter API tokens
pytest tests/e2e_gold_phase_test.py       # Full Gmail → Odoo → WhatsApp flow

---

🌱 Self-Evolution Engine

One of the most unique features of Digital-FTE is its Guardian system:

1. If any script fails, the traceback is automatically captured
2. The Guardian sends the error + context to the AI
3. The AI writes a patch and proposes it in the Vault
4. You approve → the patch is applied automatically

This means the system gets smarter and more stable over time with minimal manual intervention.

---

🗺️ Roadmap

• [ ] WhatsApp Business API (replacing Web scraping)
• [ ] Voice interface integration
• [ ] Multi-user support (team FTE)
• [ ] LangFuse observability dashboard
• [ ] Mobile app for approvals (React Native)
• [ ] Plugin marketplace for custom MCP connectors

---

🤝 Contributing

Contributions are very welcome! This is an actively maintained open-source project.

1. Fork the repo
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push and open a Pull Request

Please open an issue first for major changes so we can discuss direction.

---

👨💻 Author

Muhammad Abdullah Athar (AbdullahMalik17) Agentic AI Developer | Full-Stack Engineer | Panaversity Agentic AI Program 📍 Bahawalpur, Pakistan

• 🌐 Portfolio
• 💼 LinkedIn
• 📧 [muhammadabdullah51700@gmail.com](mailto:muhammadabdullah51700@gmail.com)

---

📄 License

This project is licensed under the MIT License.

---

⭐ If this project inspires you or saves you time, please give it a star — it helps more developers discover it!

Source & license

This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: AbdullahMalik17
• Source: AbdullahMalik17/Digital-FTE
• License: MIT

Install and usage instructions live in the source repository linked above.