copilot-proxy 0.1.2__tar.gz

copilot_proxy-0.1.2/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 modpotato
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

copilot_proxy-0.1.2/PKG-INFO

@@ -0,0 +1,188 @@
+Metadata-Version: 2.4
+Name: copilot-proxy
+Version: 0.1.2
+Summary: GitHub Copilot-compatible proxy for Z.AI GLM coding models
+Author: modpotato
+License-Expression: MIT
+Project-URL: Homepage, https://github.com/modpotato/copilot-proxy
+Project-URL: Repository, https://github.com/modpotato/copilot-proxy
+Project-URL: Issues, https://github.com/modpotato/copilot-proxy/issues
+Keywords: copilot,proxy,fastapi,glm,ollama
+Classifier: Development Status :: 4 - Beta
+Classifier: Framework :: FastAPI
+Classifier: Intended Audience :: Developers
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Internet :: Proxy Servers
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: fastapi>=0.110
+Requires-Dist: httpx>=0.27
+Requires-Dist: uvicorn[standard]>=0.27
+Requires-Dist: openai>=1.0.0
+Dynamic: license-file
+
+# Copilot-Proxy
+
+A proxy server that bridges GitHub Copilot Chat with GLM coding models by mimicking the Ollama API interface.
+
+## What it does
+
+This proxy server intercepts requests from GitHub Copilot's Ollama provider and forwards them to a GLM coding plan backend. By implementing the Ollama API interface, it allows the GitHub Copilot VS Code extension to communicate with alternative language models seamlessly.
+
+```mermaid
+flowchart TD
+   A[GitHub Copilot Chat] -- Ollama API (localhost:11434) --> B[Copilot-Proxy Server]
+   B --> C[GLM Coding Plan Backend]
+```
+
+## Quick Start
+
+### Prerequisites
+
+1. **Python 3.10+**
+2. **UV** for dependency management and packaging ([install instructions](https://docs.astral.sh/uv/getting-started/installation/))
+3. **Z.AI Coding Plan access** with a valid API key
+
+### Install from PyPI (recommended)
+
+```powershell
+# Ensure uv is installed first
+uv pip install copilot-proxy
+
+# Or run without installing globally
+uvx copilot-proxy --help
+```
+
+### Run the proxy locally
+
+```powershell
+# Quick one-liner using uvx
+uvx copilot-proxy --host 127.0.0.1 --port 11434
+
+# Or inside a synced project environment
+uv sync
+uv run copilot-proxy
+```
+
+The server listens on `http://localhost:11434` by default (same port Ollama uses). Make sure Ollama itself is stopped to avoid port conflicts.
+
+### Configure credentials
+
+Provide your Z.AI API key before launching the proxy:
+   ```powershell
+   # PowerShell (current session only)
+   $env:ZAI_API_KEY = "your-zai-api-key"
+   ```
+
+   ```bash
+   # bash/zsh
+   export ZAI_API_KEY="your-zai-api-key"
+   ```
+
+   You can optionally set a custom endpoint with `ZAI_API_BASE_URL`, though the default already targets the Coding Plan URL `https://api.z.ai/api/coding/paas/v4`.
+
+### Configure GitHub Copilot in VS Code
+
+   - Open the GitHub Copilot Chat panel in VS Code
+   - Click on the current model name to view available models
+   - Click **'Manage Models...'**
+   - Select **'Ollama'** from the list of providers
+   - Choose your preferred model from the available GLM models
+
+### Available Models
+
+The proxy advertises the GLM Coding Plan lineup so Copilot (or any Ollama-compatible client) can switch between them seamlessly:
+
+| Model        | Description                               | Use Case Highlights                          |
+|--------------|--------------------------------------------|----------------------------------------------|
+| `GLM-4.6`    | Flagship coding model with top-tier reasoning | Complex refactors, multi-file tasks, tool use |
+| `GLM-4.5`    | Balanced performance for everyday coding      | General coding, debugging, architecture input |
+| `GLM-4.5-Air`| Lightweight, faster response variant          | Quick iterations, drafting, lower-latency use |
+
+> **Tip:** These identifiers match the GLM Coding Plan catalog, so any OpenAI-compatible tool can use them by pointing to `https://api.z.ai/api/coding/paas/v4` with your Coding Plan API key.
+
+## How it Works
+
+The proxy server implements the Ollama API specification, allowing GitHub Copilot's Ollama provider to communicate with it. When Copilot sends requests to `localhost:11434`, the proxy intercepts these requests and forwards them to the GLM coding plan backend, then returns the responses in Ollama-compatible format.
+
+## Troubleshooting
+
+**Common Issues:**
+
+1. **Port conflict errors**
+   - Ensure Ollama is not running (both services use port 11434)
+   - Check that no other service is using port 11434
+   - On Windows, use: `netstat -ano | findstr :11434`
+   - On Unix/Linux/Mac, use: `lsof -i :11434`
+
+2. **Ollama provider not responding in Copilot Chat**
+   - Verify the proxy server is running
+   - Check the terminal for any error messages
+   - Ensure the GLM backend is accessible
+
+3. **Models not appearing in VS Code**
+   - Restart VS Code after starting the proxy server
+   - Make sure you've selected 'Ollama' as the provider in Copilot settings
+   - Check that the proxy server is responding at `http://localhost:11434`
+
+## Developing locally
+
+```powershell
+uv sync
+uv run uvicorn copilot_proxy.app:app --reload --port 11434
+```
+
+Use `uv run pytest` (once tests are added) or `uvx ruff check .` for linting.
+
+## Releasing to PyPI with UV
+
+1. Bump the version in `pyproject.toml`.
+2. Build the distributions:
+
+   ```powershell
+   uv build
+   ```
+
+3. Check the metadata:
+
+   ```powershell
+   uvx twine check dist/*
+   ```
+
+4. Publish to TestPyPI (recommended before production):
+
+   ```powershell
+   uv publish --repository testpypi
+   ```
+
+5. Publish to PyPI:
+
+   ```powershell
+   uv publish
+   ```
+
+Both `uv publish` commands expect the relevant API token to be available in the `UV_PUBLISH_TOKEN` environment variable.
+
+### GitHub Actions trusted publisher
+
+This repository includes `.github/workflows/publish.yml`, which builds and uploads releases automatically on GitHub tag releases. To enable it:
+
+1. Create a PyPI trusted publisher (pending or project-specific) pointing at:
+   - **Project**: `copilot-proxy`
+   - **Owner**: `modpotato`
+   - **Repository**: `copilot-proxy`
+   - **Workflow**: `publish.yml`
+   - **Environment**: `release`
+2. In GitHub, create the matching repository environment (`Settings → Environments → New environment → release`).
+3. Push a tag (e.g. `v0.1.0`) to GitHub (`git push origin v0.1.0`). The workflow will build with `uv`, publish to PyPI via OIDC, and create the GitHub release automatically.
+4. For dry runs, use the **Run workflow** button; the manual dispatch builds and validates without publishing or creating a release.
+
+## License
+
+This project is licensed under the MIT License - see the LICENSE file for details.

copilot_proxy-0.1.2/README.md

@@ -0,0 +1,159 @@
+# Copilot-Proxy
+
+A proxy server that bridges GitHub Copilot Chat with GLM coding models by mimicking the Ollama API interface.
+
+## What it does
+
+This proxy server intercepts requests from GitHub Copilot's Ollama provider and forwards them to a GLM coding plan backend. By implementing the Ollama API interface, it allows the GitHub Copilot VS Code extension to communicate with alternative language models seamlessly.
+
+```mermaid
+flowchart TD
+   A[GitHub Copilot Chat] -- Ollama API (localhost:11434) --> B[Copilot-Proxy Server]
+   B --> C[GLM Coding Plan Backend]
+```
+
+## Quick Start
+
+### Prerequisites
+
+1. **Python 3.10+**
+2. **UV** for dependency management and packaging ([install instructions](https://docs.astral.sh/uv/getting-started/installation/))
+3. **Z.AI Coding Plan access** with a valid API key
+
+### Install from PyPI (recommended)
+
+```powershell
+# Ensure uv is installed first
+uv pip install copilot-proxy
+
+# Or run without installing globally
+uvx copilot-proxy --help
+```
+
+### Run the proxy locally
+
+```powershell
+# Quick one-liner using uvx
+uvx copilot-proxy --host 127.0.0.1 --port 11434
+
+# Or inside a synced project environment
+uv sync
+uv run copilot-proxy
+```
+
+The server listens on `http://localhost:11434` by default (same port Ollama uses). Make sure Ollama itself is stopped to avoid port conflicts.
+
+### Configure credentials
+
+Provide your Z.AI API key before launching the proxy:
+   ```powershell
+   # PowerShell (current session only)
+   $env:ZAI_API_KEY = "your-zai-api-key"
+   ```
+
+   ```bash
+   # bash/zsh
+   export ZAI_API_KEY="your-zai-api-key"
+   ```
+
+   You can optionally set a custom endpoint with `ZAI_API_BASE_URL`, though the default already targets the Coding Plan URL `https://api.z.ai/api/coding/paas/v4`.
+
+### Configure GitHub Copilot in VS Code
+
+   - Open the GitHub Copilot Chat panel in VS Code
+   - Click on the current model name to view available models
+   - Click **'Manage Models...'**
+   - Select **'Ollama'** from the list of providers
+   - Choose your preferred model from the available GLM models
+
+### Available Models
+
+The proxy advertises the GLM Coding Plan lineup so Copilot (or any Ollama-compatible client) can switch between them seamlessly:
+
+| Model        | Description                               | Use Case Highlights                          |
+|--------------|--------------------------------------------|----------------------------------------------|
+| `GLM-4.6`    | Flagship coding model with top-tier reasoning | Complex refactors, multi-file tasks, tool use |
+| `GLM-4.5`    | Balanced performance for everyday coding      | General coding, debugging, architecture input |
+| `GLM-4.5-Air`| Lightweight, faster response variant          | Quick iterations, drafting, lower-latency use |
+
+> **Tip:** These identifiers match the GLM Coding Plan catalog, so any OpenAI-compatible tool can use them by pointing to `https://api.z.ai/api/coding/paas/v4` with your Coding Plan API key.
+
+## How it Works
+
+The proxy server implements the Ollama API specification, allowing GitHub Copilot's Ollama provider to communicate with it. When Copilot sends requests to `localhost:11434`, the proxy intercepts these requests and forwards them to the GLM coding plan backend, then returns the responses in Ollama-compatible format.
+
+## Troubleshooting
+
+**Common Issues:**
+
+1. **Port conflict errors**
+   - Ensure Ollama is not running (both services use port 11434)
+   - Check that no other service is using port 11434
+   - On Windows, use: `netstat -ano | findstr :11434`
+   - On Unix/Linux/Mac, use: `lsof -i :11434`
+
+2. **Ollama provider not responding in Copilot Chat**
+   - Verify the proxy server is running
+   - Check the terminal for any error messages
+   - Ensure the GLM backend is accessible
+
+3. **Models not appearing in VS Code**
+   - Restart VS Code after starting the proxy server
+   - Make sure you've selected 'Ollama' as the provider in Copilot settings
+   - Check that the proxy server is responding at `http://localhost:11434`
+
+## Developing locally
+
+```powershell
+uv sync
+uv run uvicorn copilot_proxy.app:app --reload --port 11434
+```
+
+Use `uv run pytest` (once tests are added) or `uvx ruff check .` for linting.
+
+## Releasing to PyPI with UV
+
+1. Bump the version in `pyproject.toml`.
+2. Build the distributions:
+
+   ```powershell
+   uv build
+   ```
+
+3. Check the metadata:
+
+   ```powershell
+   uvx twine check dist/*
+   ```
+
+4. Publish to TestPyPI (recommended before production):
+
+   ```powershell
+   uv publish --repository testpypi
+   ```
+
+5. Publish to PyPI:
+
+   ```powershell
+   uv publish
+   ```
+
+Both `uv publish` commands expect the relevant API token to be available in the `UV_PUBLISH_TOKEN` environment variable.
+
+### GitHub Actions trusted publisher
+
+This repository includes `.github/workflows/publish.yml`, which builds and uploads releases automatically on GitHub tag releases. To enable it:
+
+1. Create a PyPI trusted publisher (pending or project-specific) pointing at:
+   - **Project**: `copilot-proxy`
+   - **Owner**: `modpotato`
+   - **Repository**: `copilot-proxy`
+   - **Workflow**: `publish.yml`
+   - **Environment**: `release`
+2. In GitHub, create the matching repository environment (`Settings → Environments → New environment → release`).
+3. Push a tag (e.g. `v0.1.0`) to GitHub (`git push origin v0.1.0`). The workflow will build with `uv`, publish to PyPI via OIDC, and create the GitHub release automatically.
+4. For dry runs, use the **Run workflow** button; the manual dispatch builds and validates without publishing or creating a release.
+
+## License
+
+This project is licensed under the MIT License - see the LICENSE file for details.

copilot_proxy-0.1.2/copilot_proxy/__init__.py

@@ -0,0 +1,13 @@
+"""Copilot proxy package surfaces the FastAPI application and CLI utilities."""
+from __future__ import annotations
+
+from importlib import metadata
+
+from .app import app, create_app
+
+__all__ = ["app", "create_app", "__version__"]
+
+try:  # pragma: no cover - fallback when package metadata missing
+    __version__ = metadata.version("copilot-proxy")
+except metadata.PackageNotFoundError:  # type: ignore[attr-defined]
+    __version__ = "0.0.0"

copilot_proxy-0.1.2/copilot_proxy/__main__.py

@@ -0,0 +1,8 @@
+"""Allow ``python -m copilot_proxy`` to run the CLI entrypoint."""
+from __future__ import annotations
+
+from .cli import main
+
+
+if __name__ == "__main__":
+    main()

copilot_proxy-0.1.2/copilot_proxy/app.py

@@ -0,0 +1,226 @@
+"""FastAPI application exposing the Copilot proxy endpoints."""
+from __future__ import annotations
+
+import os
+from contextlib import asynccontextmanager
+from typing import AsyncGenerator
+
+import httpx
+from fastapi import FastAPI, Request
+from fastapi.responses import StreamingResponse
+
+DEFAULT_BASE_URL = "https://api.z.ai/api/coding/paas/v4"
+DEFAULT_MODEL = "GLM-4.6"
+API_KEY_ENV_VARS = ("ZAI_API_KEY", "ZAI_CODING_API_KEY", "GLM_API_KEY")
+BASE_URL_ENV_VAR = "ZAI_API_BASE_URL"
+CHAT_COMPLETION_PATH = "/chat/completions"
+
+MODEL_CATALOG = [
+    {
+        "name": "GLM-4.6",
+        "model": "GLM-4.6",
+        "modified_at": "2024-01-01T00:00:00Z",
+        "size": 0,
+        "digest": "GLM-4.6",
+        "details": {
+            "format": "glm",
+            "family": "glm",
+            "families": ["glm"],
+            "parameter_size": "cloud",
+            "quantization_level": "cloud",
+        },
+    },
+    {
+        "name": "GLM-4.5",
+        "model": "GLM-4.5",
+        "modified_at": "2024-01-01T00:00:00Z",
+        "size": 0,
+        "digest": "GLM-4.5",
+        "details": {
+            "format": "glm",
+            "family": "glm",
+            "families": ["glm"],
+            "parameter_size": "cloud",
+            "quantization_level": "cloud",
+        },
+    },
+    {
+        "name": "GLM-4.5-Air",
+        "model": "GLM-4.5-Air",
+        "modified_at": "2024-01-01T00:00:00Z",
+        "size": 0,
+        "digest": "GLM-4.5-Air",
+        "details": {
+            "format": "glm",
+            "family": "glm",
+            "families": ["glm"],
+            "parameter_size": "cloud",
+            "quantization_level": "cloud",
+        },
+    },
+]
+
+
+def _get_api_key() -> str:
+    for env_var in API_KEY_ENV_VARS:
+        api_key = os.getenv(env_var)
+        if api_key:
+            return api_key.strip()
+    raise RuntimeError(
+        "Missing Z.AI API key. Please set one of the following environment variables: "
+        + ", ".join(API_KEY_ENV_VARS)
+    )
+
+
+def _get_base_url() -> str:
+    base_url = os.getenv(BASE_URL_ENV_VAR, DEFAULT_BASE_URL).strip()
+    if not base_url:
+        base_url = DEFAULT_BASE_URL
+    if not base_url.startswith("http://") and not base_url.startswith("https://"):
+        base_url = f"https://{base_url}"
+    return base_url.rstrip("/")
+
+
+def _get_chat_completion_url() -> str:
+    base_url = _get_base_url()
+    if base_url.endswith(CHAT_COMPLETION_PATH):
+        return base_url
+    return f"{base_url}{CHAT_COMPLETION_PATH}"
+
+
+@asynccontextmanager
+async def _lifespan(app: FastAPI):  # noqa: D401 - FastAPI lifespan signature
+    """Ensure configuration is ready before serving requests."""
+
+    try:
+        _ = _get_api_key()
+        print("GLM Coding Plan proxy is ready.")
+    except Exception as exc:  # pragma: no cover - startup logging
+        print(f"Failed to initialise GLM Coding Plan proxy: {exc}")
+    yield
+
+
+def create_app() -> FastAPI:
+    """Create and return a configured FastAPI application."""
+
+    app = FastAPI(lifespan=_lifespan)
+
+    @app.get("/")
+    async def root():  # noqa: D401 - FastAPI route
+        """Return a simple health message."""
+
+        return {"message": "GLM Coding Plan proxy is running"}
+
+    @app.get("/api/ps")
+    async def list_running_models():  # noqa: D401 - FastAPI route
+        """Return an empty list as we do not host local models."""
+
+        return {"models": []}
+
+    @app.get("/api/version")
+    async def get_version():  # noqa: D401 - FastAPI route
+        """Expose a version compatible with the Ollama API expectations."""
+
+        return {"version": "0.6.4"}
+
+    @app.get("/api/tags")
+    @app.get("/api/list")
+    async def list_models():  # noqa: D401 - FastAPI route
+        """Return the static catalog of GLM models."""
+
+        return {"models": MODEL_CATALOG}
+
+    @app.post("/api/show")
+    async def show_model(request: Request):  # noqa: D401 - FastAPI route
+        """Handle Ollama-compatible model detail queries."""
+
+        try:
+            body = await request.json()
+            model_name = body.get("model")
+        except Exception:
+            model_name = DEFAULT_MODEL
+
+        if not model_name:
+            model_name = DEFAULT_MODEL
+
+        return {
+            "template": "{{ .System }}\n{{ .Prompt }}",
+            "capabilities": ["tools"],
+            "details": {
+                "family": "glm",
+                "families": ["glm"],
+                "format": "glm",
+                "parameter_size": "cloud",
+                "quantization_level": "cloud",
+            },
+            "model_info": {
+                "general.basename": model_name,
+                "general.architecture": "glm",
+                "glm.context_length": 32768,
+            },
+        }
+
+    @app.post("/v1/chat/completions")
+    async def chat_completions(request: Request):  # noqa: D401 - FastAPI route
+        """Forward chat completion calls to the Z.AI backend."""
+
+        body = await request.json()
+
+        if not body.get("model"):
+            body["model"] = DEFAULT_MODEL
+
+        stream = body.get("stream", False)
+
+        api_key = _get_api_key()
+        chat_completion_url = _get_chat_completion_url()
+
+        async def generate_chunks() -> AsyncGenerator[bytes, None]:
+            async with httpx.AsyncClient(timeout=300.0) as client:
+                try:
+                    headers = {
+                        "Content-Type": "application/json",
+                        "Authorization": f"Bearer {api_key}",
+                    }
+                    response = await client.post(
+                        chat_completion_url,
+                        headers=headers,
+                        json=body,
+                        timeout=None,
+                    )
+                    response.raise_for_status()
+
+                    async for chunk in response.aiter_bytes():
+                        yield chunk
+
+                except httpx.HTTPStatusError as exc:
+                    if exc.response.status_code == 401:
+                        raise RuntimeError("Unauthorized. Check your Z.AI API key.") from exc
+                    raise
+
+        if stream:
+            return StreamingResponse(generate_chunks(), media_type="text/event-stream")
+
+        async with httpx.AsyncClient(timeout=300.0) as client:
+            headers = {
+                "Content-Type": "application/json",
+                "Authorization": f"Bearer {api_key}",
+            }
+            response = await client.post(chat_completion_url, headers=headers, json=body)
+            response.raise_for_status()
+            return response.json()
+
+    return app
+
+
+app = create_app()
+
+__all__ = [
+    "API_KEY_ENV_VARS",
+    "BASE_URL_ENV_VAR",
+    "CHAT_COMPLETION_PATH",
+    "DEFAULT_BASE_URL",
+    "DEFAULT_MODEL",
+    "MODEL_CATALOG",
+    "app",
+    "create_app",
+]

copilot_proxy-0.1.2/copilot_proxy/cli.py

@@ -0,0 +1,61 @@
+"""Command-line interface for running the Copilot proxy."""
+from __future__ import annotations
+
+import argparse
+from typing import Optional
+
+import uvicorn
+
+DEFAULT_HOST = "127.0.0.1"
+DEFAULT_PORT = 11434
+
+
+def build_parser() -> argparse.ArgumentParser:
+    parser = argparse.ArgumentParser(description="Run the Copilot GLM proxy server.")
+    parser.add_argument(
+        "--host",
+        default=DEFAULT_HOST,
+        help=f"Host interface to bind (default: {DEFAULT_HOST}).",
+    )
+    parser.add_argument(
+        "--port",
+        type=int,
+        default=DEFAULT_PORT,
+        help=f"Port to bind (default: {DEFAULT_PORT}).",
+    )
+    parser.add_argument(
+        "--reload",
+        action="store_true",
+        help="Enable auto-reload (useful for development).",
+    )
+    parser.add_argument(
+        "--log-level",
+        default="info",
+        help="Log level passed to Uvicorn (default: info).",
+    )
+    parser.add_argument(
+        "--proxy-app",
+        default="copilot_proxy.app:app",
+        help=(
+            "Dotted path to the FastAPI application passed to Uvicorn "
+            "(default: copilot_proxy.app:app)."
+        ),
+    )
+    return parser
+
+
+def main(argv: Optional[list[str]] = None) -> None:
+    parser = build_parser()
+    args = parser.parse_args(argv)
+
+    uvicorn.run(
+        args.proxy_app,
+        host=args.host,
+        port=args.port,
+        reload=args.reload,
+        log_level=args.log_level,
+    )
+
+
+if __name__ == "__main__":
+    main()

copilot_proxy-0.1.2/copilot_proxy.egg-info/PKG-INFO

@@ -0,0 +1,188 @@
+Metadata-Version: 2.4
+Name: copilot-proxy
+Version: 0.1.2
+Summary: GitHub Copilot-compatible proxy for Z.AI GLM coding models
+Author: modpotato
+License-Expression: MIT
+Project-URL: Homepage, https://github.com/modpotato/copilot-proxy
+Project-URL: Repository, https://github.com/modpotato/copilot-proxy
+Project-URL: Issues, https://github.com/modpotato/copilot-proxy/issues
+Keywords: copilot,proxy,fastapi,glm,ollama
+Classifier: Development Status :: 4 - Beta
+Classifier: Framework :: FastAPI
+Classifier: Intended Audience :: Developers
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Internet :: Proxy Servers
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: fastapi>=0.110
+Requires-Dist: httpx>=0.27
+Requires-Dist: uvicorn[standard]>=0.27
+Requires-Dist: openai>=1.0.0
+Dynamic: license-file
+
+# Copilot-Proxy
+
+A proxy server that bridges GitHub Copilot Chat with GLM coding models by mimicking the Ollama API interface.
+
+## What it does
+
+This proxy server intercepts requests from GitHub Copilot's Ollama provider and forwards them to a GLM coding plan backend. By implementing the Ollama API interface, it allows the GitHub Copilot VS Code extension to communicate with alternative language models seamlessly.
+
+```mermaid
+flowchart TD
+   A[GitHub Copilot Chat] -- Ollama API (localhost:11434) --> B[Copilot-Proxy Server]
+   B --> C[GLM Coding Plan Backend]
+```
+
+## Quick Start
+
+### Prerequisites
+
+1. **Python 3.10+**
+2. **UV** for dependency management and packaging ([install instructions](https://docs.astral.sh/uv/getting-started/installation/))
+3. **Z.AI Coding Plan access** with a valid API key
+
+### Install from PyPI (recommended)
+
+```powershell
+# Ensure uv is installed first
+uv pip install copilot-proxy
+
+# Or run without installing globally
+uvx copilot-proxy --help
+```
+
+### Run the proxy locally
+
+```powershell
+# Quick one-liner using uvx
+uvx copilot-proxy --host 127.0.0.1 --port 11434
+
+# Or inside a synced project environment
+uv sync
+uv run copilot-proxy
+```
+
+The server listens on `http://localhost:11434` by default (same port Ollama uses). Make sure Ollama itself is stopped to avoid port conflicts.
+
+### Configure credentials
+
+Provide your Z.AI API key before launching the proxy:
+   ```powershell
+   # PowerShell (current session only)
+   $env:ZAI_API_KEY = "your-zai-api-key"
+   ```
+
+   ```bash
+   # bash/zsh
+   export ZAI_API_KEY="your-zai-api-key"
+   ```
+
+   You can optionally set a custom endpoint with `ZAI_API_BASE_URL`, though the default already targets the Coding Plan URL `https://api.z.ai/api/coding/paas/v4`.
+
+### Configure GitHub Copilot in VS Code
+
+   - Open the GitHub Copilot Chat panel in VS Code
+   - Click on the current model name to view available models
+   - Click **'Manage Models...'**
+   - Select **'Ollama'** from the list of providers
+   - Choose your preferred model from the available GLM models
+
+### Available Models
+
+The proxy advertises the GLM Coding Plan lineup so Copilot (or any Ollama-compatible client) can switch between them seamlessly:
+
+| Model        | Description                               | Use Case Highlights                          |
+|--------------|--------------------------------------------|----------------------------------------------|
+| `GLM-4.6`    | Flagship coding model with top-tier reasoning | Complex refactors, multi-file tasks, tool use |
+| `GLM-4.5`    | Balanced performance for everyday coding      | General coding, debugging, architecture input |
+| `GLM-4.5-Air`| Lightweight, faster response variant          | Quick iterations, drafting, lower-latency use |
+
+> **Tip:** These identifiers match the GLM Coding Plan catalog, so any OpenAI-compatible tool can use them by pointing to `https://api.z.ai/api/coding/paas/v4` with your Coding Plan API key.
+
+## How it Works
+
+The proxy server implements the Ollama API specification, allowing GitHub Copilot's Ollama provider to communicate with it. When Copilot sends requests to `localhost:11434`, the proxy intercepts these requests and forwards them to the GLM coding plan backend, then returns the responses in Ollama-compatible format.
+
+## Troubleshooting
+
+**Common Issues:**
+
+1. **Port conflict errors**
+   - Ensure Ollama is not running (both services use port 11434)
+   - Check that no other service is using port 11434
+   - On Windows, use: `netstat -ano | findstr :11434`
+   - On Unix/Linux/Mac, use: `lsof -i :11434`
+
+2. **Ollama provider not responding in Copilot Chat**
+   - Verify the proxy server is running
+   - Check the terminal for any error messages
+   - Ensure the GLM backend is accessible
+
+3. **Models not appearing in VS Code**
+   - Restart VS Code after starting the proxy server
+   - Make sure you've selected 'Ollama' as the provider in Copilot settings
+   - Check that the proxy server is responding at `http://localhost:11434`
+
+## Developing locally
+
+```powershell
+uv sync
+uv run uvicorn copilot_proxy.app:app --reload --port 11434
+```
+
+Use `uv run pytest` (once tests are added) or `uvx ruff check .` for linting.
+
+## Releasing to PyPI with UV
+
+1. Bump the version in `pyproject.toml`.
+2. Build the distributions:
+
+   ```powershell
+   uv build
+   ```
+
+3. Check the metadata:
+
+   ```powershell
+   uvx twine check dist/*
+   ```
+
+4. Publish to TestPyPI (recommended before production):
+
+   ```powershell
+   uv publish --repository testpypi
+   ```
+
+5. Publish to PyPI:
+
+   ```powershell
+   uv publish
+   ```
+
+Both `uv publish` commands expect the relevant API token to be available in the `UV_PUBLISH_TOKEN` environment variable.
+
+### GitHub Actions trusted publisher
+
+This repository includes `.github/workflows/publish.yml`, which builds and uploads releases automatically on GitHub tag releases. To enable it:
+
+1. Create a PyPI trusted publisher (pending or project-specific) pointing at:
+   - **Project**: `copilot-proxy`
+   - **Owner**: `modpotato`
+   - **Repository**: `copilot-proxy`
+   - **Workflow**: `publish.yml`
+   - **Environment**: `release`
+2. In GitHub, create the matching repository environment (`Settings → Environments → New environment → release`).
+3. Push a tag (e.g. `v0.1.0`) to GitHub (`git push origin v0.1.0`). The workflow will build with `uv`, publish to PyPI via OIDC, and create the GitHub release automatically.
+4. For dry runs, use the **Run workflow** button; the manual dispatch builds and validates without publishing or creating a release.
+
+## License
+
+This project is licensed under the MIT License - see the LICENSE file for details.

copilot_proxy-0.1.2/copilot_proxy.egg-info/SOURCES.txt

@@ -0,0 +1,13 @@
+LICENSE
+README.md
+pyproject.toml
+copilot_proxy/__init__.py
+copilot_proxy/__main__.py
+copilot_proxy/app.py
+copilot_proxy/cli.py
+copilot_proxy.egg-info/PKG-INFO
+copilot_proxy.egg-info/SOURCES.txt
+copilot_proxy.egg-info/dependency_links.txt
+copilot_proxy.egg-info/entry_points.txt
+copilot_proxy.egg-info/requires.txt
+copilot_proxy.egg-info/top_level.txt

copilot_proxy-0.1.2/copilot_proxy.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

copilot_proxy-0.1.2/copilot_proxy.egg-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+copilot-proxy = copilot_proxy.cli:main

copilot_proxy-0.1.2/copilot_proxy.egg-info/requires.txt

@@ -0,0 +1,4 @@
+fastapi>=0.110
+httpx>=0.27
+uvicorn[standard]>=0.27
+openai>=1.0.0

copilot_proxy-0.1.2/copilot_proxy.egg-info/top_level.txt

@@ -0,0 +1 @@
+copilot_proxy

copilot_proxy-0.1.2/pyproject.toml

@@ -0,0 +1,45 @@
+[project]
+name = "copilot-proxy"
+version = "0.1.2"
+description = "GitHub Copilot-compatible proxy for Z.AI GLM coding models"
+readme = "README.md"
+requires-python = ">=3.10"
+license = "MIT"
+authors = [{ name = "modpotato" }]
+keywords = ["copilot", "proxy", "fastapi", "glm", "ollama"]
+classifiers = [
+	"Development Status :: 4 - Beta",
+	"Framework :: FastAPI",
+	"Intended Audience :: Developers",
+	"Operating System :: OS Independent",
+	"Programming Language :: Python",
+	"Programming Language :: Python :: 3",
+	"Programming Language :: Python :: 3.10",
+	"Programming Language :: Python :: 3.11",
+	"Programming Language :: Python :: 3.12",
+	"Topic :: Internet :: Proxy Servers",
+]
+dependencies = [
+	"fastapi>=0.110",
+	"httpx>=0.27",
+	"uvicorn[standard]>=0.27",
+	"openai>=1.0.0",
+]
+
+[project.scripts]
+copilot-proxy = "copilot_proxy.cli:main"
+
+[project.urls]
+Homepage = "https://github.com/modpotato/copilot-proxy"
+Repository = "https://github.com/modpotato/copilot-proxy"
+Issues = "https://github.com/modpotato/copilot-proxy/issues"
+
+[build-system]
+requires = ["setuptools>=61.0", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[tool.setuptools.packages.find]
+include = ["copilot_proxy"]
+
+[tool.uv]
+managed = true

copilot_proxy-0.1.2/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+