ToolFront

Build AI Applications in Markdown

---

Documentation: docs.toolfront.ai

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

---

Quickstart

ToolFront is a declarative framework for building AI agents in Markdown.

Write tools and instructions in .md files. Run the project and get a live AI application.

Create it

Start with one file: README.md

---
tools:
  - [curl, -X, GET, "https://httpbin.org/status/200"]
---

# Status Checker
- Use `curl` to check if the service is up

Run it

Run the application with:

toolfront run .

Ask it

Ask your agents about the application:

Python SDK

from toolfront import Application

app = Application(url="http://127.0.0.1:8000")

result = app.ask("Is the service up?", model="openai:gpt-5")

print(result)
# Answer: yes

MCP Server

{
  "mcpServers": {
    "toolfront": {
      "command": "uvx",
      "args": ["toolfront", "mcp", "http://127.0.0.1:8000"]
    }
  }
}

---

Upgraded Example

Your full project can grow like this:

project/
├── README.md
├── src/
│   ├── api.md
│   ├── rag.md
│   ├── text2sql.md
│   └── toolkit.md
├── data/
└── tools/

Add Navigation

Update README.md with tools to explore the project

---
tools:
  - [curl, -X, GET, "https://httpbin.org/status/200"]
  - [ls]
  - [cat]
---

# Status Checker
- Use `curl` to check if the service is up
- Use `ls` and `cat` to browse other files

Add Document RAG

Give your agent tools to search documents

---
tools:
  - [grep]
---

# Search Docs
- Use `grep` to search files in `/data/`

Add Text-to-SQL

Connect your databases for SQL workflows

---
tools:
  - [psql, -U, $USER, -d, $DATABASE, -c, {query}]
---

# Database Access
- Call the `psql` tool to query the PostgreSQL database

Add Custom Tools

Build custom tools in any programming language

---
tools:
  - [python, tools/status.py, --delayed]
---

# Custom Tools
- Run `status.py` to check delayed orders

---

Installation

Install toolfront with your favorite PyPI package manager.

pip install toolfront

---

Deploy with ToolFront Cloud

Deploy your AI applications with ToolFront Cloud.

toolfront deploy ./path/to/project

This gives you a secure application URL you can access from anywhere.

app = Application(url="https://cloud.toolfront.ai/user/project", params={"API_KEY": ...})

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

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.