kiro-gateway/README.md

<div align="center">

# 👻 Kiro Gateway

**Proxy gateway for Kiro API (Amazon Q Developer / AWS CodeWhisperer)**

[🇷🇺 Русский](docs/ru/README.md) • [🇨🇳 中文](docs/zh/README.md) • [🇪🇸 Español](docs/es/README.md) • [🇮🇩 Indonesia](docs/id/README.md) • [🇧🇷 Português](docs/pt/README.md) • [🇯🇵 日本語](docs/ja/README.md) • [🇰🇷 한국어](docs/ko/README.md)

Made with ❤️ by [@Jwadow](https://github.com/jwadow)

[![License: AGPL v3](https://img.shields.io/badge/License-AGPL%20v3-blue.svg)](https://www.gnu.org/licenses/agpl-3.0)
[![Python 3.10+](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/downloads/)
[![FastAPI](https://img.shields.io/badge/FastAPI-0.100+-green.svg)](https://fastapi.tiangolo.com/)
[![Sponsor](https://img.shields.io/badge/💖_Sponsor-Support_Development-ff69b4)](#-support-the-project)

*Use Claude models from Kiro with Claude Code, OpenCode, Cursor, Cline, Roo Code, Kilo Code, Obsidian, OpenAI SDK, LangChain, Continue and other OpenAI or Anthropic compatible tools*

[Models](#-supported-models) • [Features](#-features) • [Quick Start](#-quick-start) • [Configuration](#%EF%B8%8F-configuration) • [💖 Sponsor](#-support-the-project)

</div>

---

## 🤖 Available Models

> ⚠️ **Important:** Model availability depends on your Kiro tier (free/paid). The gateway provides access to whatever models are available in your IDE or CLI based on your subscription. The list below shows models commonly available on the **free tier**.

> 🔒 **Claude Opus 4.5** was removed from the free tier on January 17, 2026. It may be available on paid tiers — check your IDE/CLI model list.

🚀 **Claude Sonnet 4.5** — Balanced performance. Great for coding, writing, and general-purpose tasks.

⚡ **Claude Haiku 4.5** — Lightning fast. Perfect for quick responses, simple tasks, and chat.

📦 **Claude Sonnet 4** — Previous generation. Still powerful and reliable for most use cases.

📦 **Claude 3.7 Sonnet** — Legacy model. Available for backward compatibility.

> 💡 **Smart Model Resolution:** Use any model name format — `claude-sonnet-4-5`, `claude-sonnet-4.5`, or even versioned names like `claude-sonnet-4-5-20250929`. The gateway normalizes them automatically.

---

## ✨ Features

| Feature | Description |
|---------|-------------|
| 🔌 **OpenAI-compatible API** | Works with any OpenAI-compatible tool |
| 🔌 **Anthropic-compatible API** | Native `/v1/messages` endpoint |
| 🌐 **VPN/Proxy Support** | HTTP/SOCKS5 proxy for restricted networks |
| 🧠 **Extended Thinking** | Reasoning is exclusive to our project |
| 👁️ **Vision Support** | Send images to model |
| 🛠️ **Tool Calling** | Supports function calling |
| 💬 **Full message history** | Passes complete conversation context |
| 📡 **Streaming** | Full SSE streaming support |
| 🔄 **Retry Logic** | Automatic retries on errors (403, 429, 5xx) |
| 📋 **Extended model list** | Including versioned models |
| 🔐 **Smart token management** | Automatic refresh before expiration |

---

## 🚀 Quick Start

### Prerequisites

- Python 3.10+
- One of the following:
  - [Kiro IDE](https://kiro.dev/) with logged in account, OR
  - [Kiro CLI](https://kiro.dev/cli/) with AWS SSO (AWS IAM Identity Center, OIDC) - free Builder ID or corporate account

### Installation

```bash
# Clone the repository (requires Git)
git clone https://github.com/Jwadow/kiro-gateway.git
cd kiro-gateway

# Or download ZIP: Code → Download ZIP → extract → open kiro-gateway folder

# Install dependencies
pip install -r requirements.txt

# Configure (see Configuration section)
cp .env.example .env
# Copy and edit .env with your credentials

# Start the server
python main.py

# Or with custom port (if 8000 is busy)
python main.py --port 9000
```

The server will be available at `http://localhost:8000`

---

## ⚙️ Configuration

### Option 1: JSON Credentials File (Kiro IDE / Enterprise)

Specify the path to the credentials file:

Works with:
- **Kiro IDE** (standard) - for personal accounts
- **Enterprise** - for corporate accounts with SSO

```env
KIRO_CREDS_FILE="~/.aws/sso/cache/kiro-auth-token.json"

# Password to protect YOUR proxy server (make up any secure string)
# You'll use this as api_key when connecting to your gateway
PROXY_API_KEY="my-super-secret-password-123"
```

<details>
<summary>📄 JSON file format</summary>

```json
{
  "accessToken": "eyJ...",
  "refreshToken": "eyJ...",
  "expiresAt": "2025-01-12T23:00:00.000Z",
  "profileArn": "arn:aws:codewhisperer:us-east-1:...",
  "region": "us-east-1",
  "clientIdHash": "abc123..."  // Optional: for corporate SSO setups
}
```

> **Note:** If you have two JSON files in `~/.aws/sso/cache/` (e.g., `kiro-auth-token.json` and a file with a hash name), use `kiro-auth-token.json` in `KIRO_CREDS_FILE`. The gateway will automatically load the other file.

</details>

### Option 2: Environment Variables (.env file)

Create a `.env` file in the project root:

```env
# Required
REFRESH_TOKEN="your_kiro_refresh_token"

# Password to protect YOUR proxy server (make up any secure string)
PROXY_API_KEY="my-super-secret-password-123"

# Optional
PROFILE_ARN="arn:aws:codewhisperer:us-east-1:..."
KIRO_REGION="us-east-1"
```

### Option 3: AWS SSO Credentials (kiro-cli / Enterprise)

If you use `kiro-cli` or Kiro IDE with AWS SSO (AWS IAM Identity Center), the gateway will automatically detect and use the appropriate authentication.

Works with both free Builder ID accounts and corporate accounts.

```env
KIRO_CREDS_FILE="~/.aws/sso/cache/your-sso-cache-file.json"

# Password to protect YOUR proxy server
PROXY_API_KEY="my-super-secret-password-123"

# Note: PROFILE_ARN is NOT needed for AWS SSO (Builder ID and corporate accounts)
# The gateway will work without it
```

<details>
<summary>📄 AWS SSO JSON file format</summary>

AWS SSO credentials files (from `~/.aws/sso/cache/`) contain:

```json
{
  "accessToken": "eyJ...",
  "refreshToken": "eyJ...",
  "expiresAt": "2025-01-12T23:00:00.000Z",
  "region": "us-east-1",
  "clientId": "...",
  "clientSecret": "..."
}
```

**Note:** AWS SSO (Builder ID and corporate accounts) users do NOT need `profileArn`. The gateway will work without it (if specified, it will be ignored).

</details>

<details>
<summary>🔍 How it works</summary>

The gateway automatically detects the authentication type based on the credentials file:

- **Kiro Desktop Auth** (default): Used when `clientId` and `clientSecret` are NOT present
  - Endpoint: `https://prod.{region}.auth.desktop.kiro.dev/refreshToken`
  
- **AWS SSO (OIDC)**: Used when `clientId` and `clientSecret` ARE present
  - Endpoint: `https://oidc.{region}.amazonaws.com/token`

No additional configuration is needed — just point to your credentials file!

</details>

### Option 4: kiro-cli SQLite Database

If you use `kiro-cli` and prefer to use its SQLite database directly:

```env
KIRO_CLI_DB_FILE="~/.local/share/kiro-cli/data.sqlite3"

# Password to protect YOUR proxy server
PROXY_API_KEY="my-super-secret-password-123"

# Note: PROFILE_ARN is NOT needed for AWS SSO (Builder ID and corporate accounts)
# The gateway will work without it
```

<details>
<summary>📄 Database locations</summary>

| CLI Tool | Database Path |
|----------|---------------|
| kiro-cli | `~/.local/share/kiro-cli/data.sqlite3` |
| amazon-q-developer-cli | `~/.local/share/amazon-q/data.sqlite3` |

The gateway reads credentials from the `auth_kv` table which stores:
- `kirocli:odic:token` or `codewhisperer:odic:token` — access token, refresh token, expiration
- `kirocli:odic:device-registration` or `codewhisperer:odic:device-registration` — client ID and secret

Both key formats are supported for compatibility with different kiro-cli versions.

</details>

### Getting Credentials

**For Kiro IDE users:**
- Log in to Kiro IDE and use Option 1 above (JSON credentials file)
- The credentials file is created automatically after login

**For Kiro CLI users:**
- Log in with `kiro-cli login` and use Option 3 or Option 4 above
- No manual token extraction needed!

<details>
<summary>🔧 Advanced: Manual token extraction</summary>

If you need to manually extract the refresh token (e.g., for debugging), you can intercept Kiro IDE traffic:
- Look for requests to: `prod.us-east-1.auth.desktop.kiro.dev/refreshToken`

</details>

---

## 🌐 VPN/Proxy Support

**For users in China, corporate networks, or regions with connectivity issues to AWS services.**

The gateway supports routing all Kiro API requests through a VPN or proxy server. This is essential if you experience connection problems to AWS endpoints or need to use a corporate proxy.

### Configuration

Add to your `.env` file:

```env
# HTTP proxy
VPN_PROXY_URL=http://127.0.0.1:7890

# SOCKS5 proxy
VPN_PROXY_URL=socks5://127.0.0.1:1080

# With authentication (corporate proxies)
VPN_PROXY_URL=http://username:password@proxy.company.com:8080

# Without protocol (defaults to http://)
VPN_PROXY_URL=192.168.1.100:8080
```

### Supported Protocols

- ✅ **HTTP** — Standard proxy protocol
- ✅ **HTTPS** — Secure proxy connections
- ✅ **SOCKS5** — Advanced proxy protocol (common in VPN software)
- ✅ **Authentication** — Username/password embedded in URL

### When You Need This

| Situation | Solution |
|-----------|----------|
| Connection timeouts to AWS | Use VPN/proxy to route traffic |
| Corporate network restrictions | Configure your company's proxy |
| Regional connectivity issues | Use a VPN service with proxy support |
| Privacy requirements | Route through your own proxy server |

### Popular VPN Software with Proxy Support

Most VPN clients provide a local proxy server you can use:
- **Sing-box** — Modern VPN client with HTTP/SOCKS5 proxy
- **Clash** — Usually runs on `http://127.0.0.1:7890`
- **V2Ray** — Configurable SOCKS5/HTTP proxy
- **Shadowsocks** — SOCKS5 proxy support
- **Corporate VPN** — Check your IT department for proxy settings

Leave `VPN_PROXY_URL` empty (default) if you don't need proxy support.

---

## 📡 API Reference

### Endpoints

| Endpoint | Method | Description |
|----------|--------|-------------|
| `/` | GET | Health check |
| `/health` | GET | Detailed health check |
| `/v1/models` | GET | List available models |
| `/v1/chat/completions` | POST | OpenAI Chat Completions API |
| `/v1/messages` | POST | Anthropic Messages API |

---

## 💡 Usage Examples

### OpenAI API

<details>
<summary>🔹 Simple cURL Request</summary>

```bash
curl http://localhost:8000/v1/chat/completions \
  -H "Authorization: Bearer my-super-secret-password-123" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-5",
    "messages": [{"role": "user", "content": "Hello!"}],
    "stream": true
  }'
```

> **Note:** Replace `my-super-secret-password-123` with the `PROXY_API_KEY` you set in your `.env` file.

</details>

<details>
<summary>🔹 Streaming Request</summary>

```bash
curl http://localhost:8000/v1/chat/completions \
  -H "Authorization: Bearer my-super-secret-password-123" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-5",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "What is 2+2?"}
    ],
    "stream": true
  }'
```

</details>

<details>
<summary>🛠️ With Tool Calling</summary>

```bash
curl http://localhost:8000/v1/chat/completions \
  -H "Authorization: Bearer my-super-secret-password-123" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-5",
    "messages": [{"role": "user", "content": "What is the weather in London?"}],
    "tools": [{
      "type": "function",
      "function": {
        "name": "get_weather",
        "description": "Get weather for a location",
        "parameters": {
          "type": "object",
          "properties": {
            "location": {"type": "string", "description": "City name"}
          },
          "required": ["location"]
        }
      }
    }]
  }'
```

</details>

<details>
<summary>🐍 Python OpenAI SDK</summary>

```python
from openai import OpenAI

client = OpenAI(
    base_url="http://localhost:8000/v1",
    api_key="my-super-secret-password-123"  # Your PROXY_API_KEY from .env
)

response = client.chat.completions.create(
    model="claude-sonnet-4-5",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello!"}
    ],
    stream=True
)

for chunk in response:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="")
```

</details>

<details>
<summary>🦜 LangChain</summary>

```python
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    base_url="http://localhost:8000/v1",
    api_key="my-super-secret-password-123",  # Your PROXY_API_KEY from .env
    model="claude-sonnet-4-5"
)

response = llm.invoke("Hello, how are you?")
print(response.content)
```

</details>

### Anthropic API

<details>
<summary>🔹 Simple cURL Request</summary>

```bash
curl http://localhost:8000/v1/messages \
  -H "x-api-key: my-super-secret-password-123" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-5",
    "max_tokens": 1024,
    "messages": [{"role": "user", "content": "Hello!"}]
  }'
```

> **Note:** Anthropic API uses `x-api-key` header instead of `Authorization: Bearer`. Both are supported.

</details>

<details>
<summary>🔹 With System Prompt</summary>

```bash
curl http://localhost:8000/v1/messages \
  -H "x-api-key: my-super-secret-password-123" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-5",
    "max_tokens": 1024,
    "system": "You are a helpful assistant.",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'
```

> **Note:** In Anthropic API, `system` is a separate field, not a message.

</details>

<details>
<summary>📡 Streaming</summary>

```bash
curl http://localhost:8000/v1/messages \
  -H "x-api-key: my-super-secret-password-123" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-5",
    "max_tokens": 1024,
    "stream": true,
    "messages": [{"role": "user", "content": "Hello!"}]
  }'
```

</details>

<details>
<summary>🐍 Python Anthropic SDK</summary>

```python
import anthropic

client = anthropic.Anthropic(
    api_key="my-super-secret-password-123",  # Your PROXY_API_KEY from .env
    base_url="http://localhost:8000"
)

# Non-streaming
response = client.messages.create(
    model="claude-sonnet-4-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.content[0].text)

# Streaming
with client.messages.stream(
    model="claude-sonnet-4-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello!"}]
) as stream:
    for text in stream.text_stream:
        print(text, end="", flush=True)
```

</details>

---

## 🔧 Debugging

Debug logging is **disabled by default**. To enable, add to your `.env`:

```env
# Debug logging mode:
# - off: disabled (default)
# - errors: save logs only for failed requests (4xx, 5xx) - recommended for troubleshooting
# - all: save logs for every request (overwrites on each request)
DEBUG_MODE=errors
```

### Debug Modes

| Mode | Description | Use Case |
|------|-------------|----------|
| `off` | Disabled (default) | Production |
| `errors` | Save logs only for failed requests (4xx, 5xx) | **Recommended for troubleshooting** |
| `all` | Save logs for every request | Development/debugging |

### Debug Files

When enabled, requests are logged to the `debug_logs/` folder:

| File | Description |
|------|-------------|
| `request_body.json` | Incoming request from client (OpenAI format) |
| `kiro_request_body.json` | Request sent to Kiro API |
| `response_stream_raw.txt` | Raw stream from Kiro |
| `response_stream_modified.txt` | Transformed stream (OpenAI format) |
| `app_logs.txt` | Application logs for the request |
| `error_info.json` | Error details (only on errors) |

---

## 📜 License

This project is licensed under the **GNU Affero General Public License v3.0 (AGPL-3.0)**.

This means:
- ✅ You can use, modify, and distribute this software
- ✅ You can use it for commercial purposes
- ⚠️ **You must disclose source code** when you distribute the software
- ⚠️ **Network use is distribution** — if you run a modified version on a server and let others interact with it, you must make the source code available to them
- ⚠️ Modifications must be released under the same license

See the [LICENSE](LICENSE) file for the full license text.

### Why AGPL-3.0?

AGPL-3.0 ensures that improvements to this software benefit the entire community. If you modify this gateway and deploy it as a service, you must share your improvements with your users.

### Contributor License Agreement (CLA)

By submitting a contribution to this project, you agree to the terms of our [Contributor License Agreement (CLA)](CLA.md). This ensures that:
- You have the right to submit the contribution
- You grant the maintainer rights to use and relicense your contribution
- The project remains legally protected

---

## 💖 Support the Project

<div align="center">

<img src="https://raw.githubusercontent.com/Tarikul-Islam-Anik/Animated-Fluent-Emojis/master/Emojis/Smilies/Smiling%20Face%20with%20Hearts.png" alt="Love" width="80" />

**If this project saved you time or money, consider supporting it!**

Every contribution helps keep this project alive and growing

<br>

### 🤑 Donate

[**☕ One-time Donation**](https://app.lava.top/jwadow?tabId=donate) &nbsp;•&nbsp; [**💎 Monthly Support**](https://app.lava.top/jwadow?tabId=subscriptions)

<br>

### 🪙 Or send crypto

| Currency | Network | Address |
|:--------:|:-------:|:--------|
| **USDT** | TRC20 | `TSVtgRc9pkC1UgcbVeijBHjFmpkYHDRu26` |
| **BTC** | Bitcoin | `12GZqxqpcBsqJ4Vf1YreLqwoMGvzBPgJq6` |
| **ETH** | Ethereum | `0xc86eab3bba3bbaf4eb5b5fff8586f1460f1fd395` |
| **SOL** | Solana | `9amykF7KibZmdaw66a1oqYJyi75fRqgdsqnG66AK3jvh` |
| **TON** | TON | `UQBVh8T1H3GI7gd7b-_PPNnxHYYxptrcCVf3qQk5v41h3QTM` |

</div>

---

## ⚠️ Disclaimer

This project is not affiliated with, endorsed by, or sponsored by Amazon Web Services (AWS), Anthropic, or Kiro IDE. Use at your own risk and in compliance with the terms of service of the underlying APIs.

---

<div align="center">

**[⬆ Back to Top](#-kiro-gateway)**

</div>