Metadata-Version: 2.4
Name: kalibr
Version: 0.1.0
Summary: Connect any API to AI platforms in 10 lines of code
Home-page: https://github.com/yourusername/kalibr-sdk
Author: Your Name
Author-email: your.email@example.com
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
Requires-Dist: requests>=2.28.0
Requires-Dist: fastapi>=0.100.0
Requires-Dist: uvicorn>=0.20.0
Requires-Dist: httpx>=0.24.0
Dynamic: author
Dynamic: author-email
Dynamic: classifier
Dynamic: description
Dynamic: description-content-type
Dynamic: home-page
Dynamic: requires-dist
Dynamic: requires-python
Dynamic: summary


# Kalibr SDK Docs

Kalibr makes your SaaS agent-ready.  
Define your API once and Kalibr compiles connectors for ChatGPT, Claude, and more.  
Our MVP works today with GPT and Claude. The universal proxy makes your API callable in minutes.

---

## Quickstart (MVP)

### Install
```bash
pip install kalibr-sdk

Your First Integration

from kalibr_sdk import Kalibr

# Connect your backend
k = Kalibr()
k.connect("https://api.airtable.com", api_key="key123")

# Register actions
k.register_tool("create_record", "/v0/base123/Table", "POST")
k.register_tool("search_records", "/v0/base123/Table", "GET")

# Deploy
k.deploy()

What Happens
	•	Kalibr spins up the universal proxy
	•	Exposes your API at public URLs via ngrok
	•	Generates platform configs (OpenAPI for GPT, MCP for Claude)
	•	You copy the configs into ChatGPT or Claude → your API works

⸻

Demo Experience

With 5–7 lines of code:
	•	Your API works with ChatGPT and Claude
	•	No platform-specific code
	•	Real requests hit your production backend

Example:
	•	ChatGPT → “Add a contact to my CRM”
	•	Claude → “Search for customers from Microsoft”

Both hit your Airtable API through Kalibr.

⸻

Current Status (MVP)

GPT Actions (OpenAPI) working
Claude MCP working
Universal proxy live (local + ngrok)
Register actions once, run on multiple models

⸻

What’s Next

With funding + partners, we’ll expand:
	•	Hosted proxy (multi-tenant, secure, SOC2)
	•	More platforms (Gemini, Azure Copilot, Assistants)
	•	Normalization layer (auth, error codes, schema drift)
	•	Drift detection (auto-patch when providers change)
	•	Auto-registration (where platforms support it)

⸻

Roadmap

MVP (today)
	•	Local proxy + SDK
	•	GPT + Claude connectors
	•	Design partner pilots

Phase 2 (3–6 months)
	•	Hosted proxy (AWS/GCP/Azure)
	•	Multi-tenant infra
	•	OAuth + advanced auth handling
	•	Gemini + Azure Copilot support

Phase 3 (12–18 months)
	•	Auto-registration
	•	Normalization at scale
	•	Drift detection + monitoring
	•	Compliance, reliability, uptime SLAs

⸻

Design Partner Program

We’re onboarding a small group of SaaS design partners to shape Kalibr.

Design partners get:
	•	Early SDK access
	•	Forward-deployed engineering support
	•	Roadmap influence
	•	Preferred pricing

📩 Contact: hello@kalibr.systems

This is clean, Markdown-friendly, and looks normal to a developer.  

Want me to also scaffold an **`examples/` folder with one Airtable example** so it looks even more real on GitHub, or do you want to keep it super minimal for now?
