kalibr 1.0.21

Multi-Model AI Integration Framework

Kalibr SDK

Multi-Model AI Integration Framework

Write once. Deploy anywhere. Connect to any AI model.

Kalibr lets you expose Python functions as APIs that work with GPT, Claude, Gemini, and Copilot automatically.

---

🚀 Quick Start (2 minutes)

1. Install

pip install kalibr

2. Get Examples

kalibr-connect examples

This copies example files to ./kalibr_examples/ in your current directory.

3. Run Demo

kalibr-connect serve kalibr_examples/basic_kalibr_example.py

4. See Multi-Model Schemas

Open your browser to see all 4 AI model schemas auto-generated:

http://localhost:8000/                    # API info
http://localhost:8000/gpt-actions.json    # For ChatGPT
http://localhost:8000/mcp.json            # For Claude
http://localhost:8000/schemas/gemini      # For Gemini
http://localhost:8000/schemas/copilot     # For Copilot
http://localhost:8000/docs                # Interactive docs

That's it! One Python file, four AI platform schemas. 🎯

---

🎯 What Does This Do?

Kalibr turns your Python functions into APIs that AI assistants can call:

from kalibr import Kalibr

app = Kalibr(title="My Business API")

@app.action("get_inventory", "Check product stock")
def get_inventory(product_id: str):
    # Your business logic
    return {"product_id": product_id, "stock": 42}

Result: ChatGPT, Claude, Gemini, and Copilot can all call your get_inventory function!

---

💪 Two Modes

Function-Level (Simple)

Perfect for exposing business logic:

from kalibr import Kalibr

app = Kalibr(title="My API")

@app.action("calculate_price", "Calculate product price")
def calculate_price(product_id: str, quantity: int):
    return {"total": quantity * get_price(product_id)}

App-Level (Advanced)

Full framework with file uploads, sessions, streaming, workflows:

from kalibr import KalibrApp
from kalibr.types import FileUpload, Session

app = KalibrApp(title="Advanced API")

@app.file_handler("analyze_doc", [".pdf", ".docx"])
async def analyze_doc(file: FileUpload):
    return {"filename": file.filename, "analysis": "..."}

@app.session_action("save_data", "Save to session")
async def save_data(session: Session, data: dict):
    session.set("my_data", data)
    return {"saved": True}

@app.stream_action("live_feed", "Stream real-time data")
async def live_feed(count: int = 10):
    for i in range(count):
        yield {"item": i, "timestamp": "..."}

---

📚 Examples Included

After running kalibr-connect examples, you get:

• basic_kalibr_example.py - Simple function-level example
• enhanced_kalibr_example.py - Advanced app-level example with all features

---

🤖 AI Platform Integration

Once your Kalibr app is running, integrate with AI platforms:

ChatGPT (GPT Actions)

1. Copy schema from http://localhost:8000/gpt-actions.json
2. Go to GPT Builder → Actions → Import from URL
3. Done! ChatGPT can now call your functions

Claude (MCP)

1. Add to Claude Desktop config:

{
  "mcp": {
    "servers": {
      "my-api": {
        "url": "http://localhost:8000/mcp.json"
      }
    }
  }
}

2. Done! Claude can now call your functions

Gemini & Copilot

Similar simple setup using their respective schema endpoints.

---

🎯 Use Cases

• Customer Service APIs - Let AI assistants look up orders, process refunds
• Data Analysis APIs - Let AI query your analytics and generate insights
• Document Processing - Let AI analyze uploaded documents
• Business Automation - Let AI trigger workflows in your systems
• Internal Tools - Give your team AI-powered access to internal systems

---

📖 Documentation

• Quick Start: You're reading it!
• Full Docs: See KALIBR_SDK_COMPLETE.md in the package
• Examples: Run kalibr-connect examples
• CLI Help: kalibr-connect --help

---

🔧 CLI Commands

kalibr-connect examples          # Copy example files to current dir
kalibr-connect serve my_app.py   # Run your app locally
kalibr-connect version           # Show version info
kalibr-connect --help            # Show all commands

---

⚡ Key Features

• ✅ Multi-Model Support - Works with GPT, Claude, Gemini, Copilot
• ✅ Automatic Schemas - No manual schema writing
• ✅ File Uploads - Handle document uploads
• ✅ Sessions - Stateful conversations
• ✅ Streaming - Real-time data streaming
• ✅ Workflows - Multi-step processes
• ✅ Type Safe - Full Python type hints
• ✅ Fast - Async/await support

---

🔥 Why Kalibr?

Without Kalibr:

• Learn 4 different API specs
• Write 4 different schemas
• Maintain 4 codebases
• = Weeks of work

With Kalibr:

• Write Python functions once
• Kalibr generates all 4 schemas
• Single codebase
• = One day of work

---

💡 Simple Example

# my_app.py
from kalibr import Kalibr

app = Kalibr(title="Weather API")

@app.action("get_weather", "Get current weather")
def get_weather(city: str):
    # Your logic here
    return {"city": city, "temp": 72, "condition": "sunny"}

# Run it
kalibr-connect serve my_app.py

# Now ALL these work:
# ✅ ChatGPT can call get_weather()
# ✅ Claude can call get_weather()
# ✅ Gemini can call get_weather()
# ✅ Copilot can call get_weather()

---

🚀 Get Started Now

pip install kalibr
kalibr-connect examples
kalibr-connect serve kalibr_examples/basic_kalibr_example.py

Open http://localhost:8000 and see your multi-model API in action! 🎯

---

📝 License

MIT License - see LICENSE file for details.

---

Kalibr SDK - Transform how you build AI-integrated applications! 🚀