ToolFront

Documentation: docs.toolfront.ai

Source code: https://github.com/statespace-tech/toolfront

Installation

Install toolfront with your favorite PyPI package manager.

pip install toolfront

Quickstart

ToolFront helps you organize AI workflows into independent tasks with data environments.

environment/
├── index.md
├── pages/
│   ├── text2sql.md
│   ├── document.md
│   └── api.md
└── data/
    ├── invoices/
    └── logs/

5 directories, 15 files

Declare task instructions and tools in Markdown files.

---
tools:
  - [date, +%Y-%m-%d]

---

# Landing Page

- Include links to [pages](./pages)
- Add tool commands to frontmatters
- Agents learn tools with `--help`

Create text-to-SQL tasks with ToolFront's built-in database CLI.

---
tools:
  - [toolfront, database, $DB_URL]

---

# Text-to-SQL

- Add database metadata and context
- Agents can list and inspect tables
- All queries are read-only

Retrieve information from data files like .txt, .csv, and .json.

---
tools:
  - [python, extract.py]

---

# Document RAG

- Add data files and descriptions
- Agents read and search documents
- Use custom tools to process data

Fetch live data with calls to external APIs.

---
tools:
  - [curl, "https://api.com/v1/user"]

---

# API Integration

- Define API endpoints as tools
- Pass env `$VARS` for secrets
- Agents fetch live external data

Agents browse environments to get work done, using tools and following instructions as needed.

Run Python agents on environments with the SDK.

from toolfront import Environment

env = Environment(url="file:///path/environment")

response = env.run("What's our average ticket size?", model="openai:gpt-5")

Connect your own agents to environments with the MCP Server.

{
  "mcpServers": {
    "toolfront": {
      "command": "uvx",
      "args": ["toolfront", "mcp", "file:///path/environment"]
    }
  }
}

Deploy with ToolFront Cloud

Instantly deploy your environments with ToolFront Cloud.

toolfront deploy ./path/environment

This gives you a secure environment URL your agents can access from anywhere.

Environment(url="https://cloud.toolfront.ai/user/environment", params={"API_KEY": ...})

ToolFront Cloud is in beta. To request access, join our Discord or email esteban[at]kruskal[dot]ai.

Community & Contributing

• Discord: Join our community server for real-time help and discussions
• X: Follow us @toolfront for updates and news
• Issues: Report bugs or request features on GitHub Issues

License

This project is licensed under the terms of the MIT license.