Skip to content

ccproxy.api.routes.proxy

ccproxy.api.routes.proxy

Proxy endpoints for CCProxy API Server.

create_openai_chat_completion async 

create_openai_chat_completion(request, proxy_service, auth)

Create a chat completion using Claude AI with OpenAI-compatible format.

This endpoint handles OpenAI API format requests and forwards them directly to Claude via the proxy service.

Source code in ccproxy/api/routes/proxy.py 
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
@router.post("/v1/chat/completions", response_model=None)
async def create_openai_chat_completion(
    request: Request,
    proxy_service: ProxyServiceDep,
    auth: ConditionalAuthDep,
) -> StreamingResponse | Response:
    """Create a chat completion using Claude AI with OpenAI-compatible format.

    This endpoint handles OpenAI API format requests and forwards them
    directly to Claude via the proxy service.
    """
    try:
        # Get request body
        body = await request.body()

        # Get headers and query params
        headers = dict(request.headers)
        query_params: dict[str, str | list[str]] | None = (
            dict(request.query_params) if request.query_params else None
        )

        # Handle the request using proxy service directly
        response = await proxy_service.handle_request(
            method=request.method,
            path=request.url.path,
            headers=headers,
            body=body,
            query_params=query_params,
            request=request,  # Pass the request object for context access
        )

        # Return appropriate response type
        if isinstance(response, StreamingResponse):
            # Already a streaming response
            return response
        else:
            # Tuple response - handle regular response
            status_code, response_headers, response_body = response
            if status_code >= 400:
                # Forward error response directly with headers
                return ProxyResponse(
                    content=response_body,
                    status_code=status_code,
                    headers=response_headers,
                    media_type=response_headers.get("content-type", "application/json"),
                )

            # Check if this is a streaming response based on content-type
            content_type = response_headers.get("content-type", "")
            if "text/event-stream" in content_type:
                # Return as streaming response
                async def stream_generator() -> AsyncIterator[bytes]:
                    # Split the SSE data into chunks
                    for line in response_body.decode().split("\n"):
                        if line.strip():
                            yield f"{line}\n".encode()

                return StreamingResponse(
                    stream_generator(),
                    media_type="text/event-stream",
                    headers={
                        "Cache-Control": "no-cache",
                        "Connection": "keep-alive",
                    },
                )
            else:
                # Parse JSON response
                response_data = json.loads(response_body.decode())

                # Convert Anthropic response back to OpenAI format for /chat/completions
                openai_adapter = OpenAIAdapter()
                openai_response = openai_adapter.adapt_response(response_data)

                # Return response with headers
                return ProxyResponse(
                    content=json.dumps(openai_response),
                    status_code=status_code,
                    headers=response_headers,
                    media_type=response_headers.get("content-type", "application/json"),
                )

    except HTTPException:
        # Re-raise HTTPException as-is (including 401 auth errors)
        raise
    except Exception as e:
        raise HTTPException(
            status_code=500, detail=f"Internal server error: {str(e)}"
        ) from e

create_anthropic_message async 

create_anthropic_message(request, proxy_service, auth)

Create a message using Claude AI with Anthropic format.

This endpoint handles Anthropic API format requests and forwards them directly to Claude via the proxy service.

Source code in ccproxy/api/routes/proxy.py 
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
@router.post("/v1/messages", response_model=None)
async def create_anthropic_message(
    request: Request,
    proxy_service: ProxyServiceDep,
    auth: ConditionalAuthDep,
) -> StreamingResponse | Response:
    """Create a message using Claude AI with Anthropic format.

    This endpoint handles Anthropic API format requests and forwards them
    directly to Claude via the proxy service.
    """
    try:
        # Get request body
        body = await request.body()

        # Get headers and query params
        headers = dict(request.headers)
        query_params: dict[str, str | list[str]] | None = (
            dict(request.query_params) if request.query_params else None
        )

        # Handle the request using proxy service directly
        response = await proxy_service.handle_request(
            method=request.method,
            path=request.url.path,
            headers=headers,
            body=body,
            query_params=query_params,
            request=request,  # Pass the request object for context access
        )

        # Return appropriate response type
        if isinstance(response, StreamingResponse):
            # Already a streaming response
            return response
        else:
            # Tuple response - handle regular response
            status_code, response_headers, response_body = response
            if status_code >= 400:
                # Forward error response directly with headers
                return ProxyResponse(
                    content=response_body,
                    status_code=status_code,
                    headers=response_headers,
                    media_type=response_headers.get("content-type", "application/json"),
                )

            # Check if this is a streaming response based on content-type
            content_type = response_headers.get("content-type", "")
            if "text/event-stream" in content_type:
                # Return as streaming response
                async def stream_generator() -> AsyncIterator[bytes]:
                    # Split the SSE data into chunks
                    for line in response_body.decode().split("\n"):
                        if line.strip():
                            yield f"{line}\n".encode()

                return StreamingResponse(
                    stream_generator(),
                    media_type="text/event-stream",
                    headers={
                        "Cache-Control": "no-cache",
                        "Connection": "keep-alive",
                    },
                )
            else:
                # Parse JSON response
                response_data = json.loads(response_body.decode())

                # Return response with headers
                return ProxyResponse(
                    content=response_body,  # Use original body to preserve exact format
                    status_code=status_code,
                    headers=response_headers,
                    media_type=response_headers.get("content-type", "application/json"),
                )

    except HTTPException:
        # Re-raise HTTPException as-is (including 401 auth errors)
        raise
    except Exception as e:
        raise HTTPException(
            status_code=500, detail=f"Internal server error: {str(e)}"
        ) from e

list_models async 

list_models(request, proxy_service, auth)

List available models using the proxy service.

Returns a combined list of Anthropic models and recent OpenAI models.

Source code in ccproxy/api/routes/proxy.py 
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
@router.get("/v1/models", response_model=None)
async def list_models(
    request: Request,
    proxy_service: ProxyServiceDep,
    auth: ConditionalAuthDep,
) -> Response:
    """List available models using the proxy service.

    Returns a combined list of Anthropic models and recent OpenAI models.
    """
    try:
        # Get headers
        headers = dict(request.headers)

        # Handle the request using proxy service
        response = await proxy_service.handle_request(
            method="GET",
            path="/v1/models",
            headers=headers,
            body=None,
            request=request,
        )

        # Since /v1/models never streams, we know it returns a tuple
        if isinstance(response, tuple):
            status_code, response_headers, response_body = response
        else:
            # This shouldn't happen for /v1/models, but handle it gracefully
            raise HTTPException(
                status_code=500,
                detail="Unexpected streaming response for /v1/models endpoint",
            )

        # Return response with headers
        return ProxyResponse(
            content=response_body,
            status_code=status_code,
            headers=response_headers,
            media_type=response_headers.get("content-type", "application/json"),
        )

    except HTTPException:
        # Re-raise HTTPException as-is (including 401 auth errors)
        raise
    except Exception as e:
        raise HTTPException(
            status_code=500, detail=f"Internal server error: {str(e)}"
        ) from e