Claude/fix GitHub issues 01 lscdh eb bpmpqn usd y1go8 x

README.md

@@ -21,6 +21,7 @@
 
 - **Model Routing**: Route requests to different models based on your needs (e.g., background tasks, thinking, long context).
 - **Multi-Provider Support**: Supports various model providers like OpenRouter, DeepSeek, Ollama, Gemini, Volcengine, and SiliconFlow.
+- **Multiple Instances**: Run unlimited CCR instances simultaneously with different configurations.
 - **Request/Response Transformation**: Customize requests and responses for different providers using transformers.
 - **Dynamic Model Switching**: Switch models on-the-fly within Claude Code using the `/model` command.
 - **CLI Model Management**: Manage models and providers directly from the terminal with `ccr model`.
@@ -413,6 +414,56 @@ The `Router` object defines which model to use for different scenarios:
 `/model provider_name,model_name`
 Example: `/model openrouter,anthropic/claude-3.5-sonnet`
 
+#### Configuration Examples
+
+This repository includes several ready-to-use configuration examples in the root directory:
+
+- **`config.example.json`** - Comprehensive example with all major providers (OpenRouter, DeepSeek, Gemini, Ollama, Groq, Copilot API)
+- **`config.copilot.example.json`** - GitHub Copilot API integration example (see below)
+- **`config.multimodel.example.json`** - Multi-model strategy for cost optimization
+
+Copy and customize these examples to `~/.claude-code-router/config.json` to get started quickly.
+
+#### GitHub Copilot API Integration
+
+Claude Code Router supports integration with [copilot-api](https://github.com/ericc-ch/copilot-api), a local API wrapper that provides access to multiple AI models through a unified interface.
+
+**Setup Steps:**
+
+1. Install and start copilot-api (default port: 4141)
+2. Configure CCR to use the Copilot API endpoint:
+
+```json
+{
+  "Providers": [
+    {
+      "name": "copilot",
+      "api_base_url": "http://localhost:4141/v1/messages",
+      "api_key": "dummy-key-for-local",
+      "models": [
+        "claude-sonnet-4.5",
+        "gpt-5.1",
+        "gpt-5.1-codex"
+      ],
+      "transformer": {
+        "use": ["anthropic"]
+      }
+    }
+  ],
+  "Router": {
+    "default": "copilot,claude-sonnet-4.5",
+    "think": "copilot,gpt-5.1"
+  }
+}
+```
+
+**Important:** Use the `anthropic` transformer for proper message format compatibility. See `config.copilot.example.json` for a complete working example.
+
+**Working Models:** Claude Sonnet 4.5, GPT-5.1, GPT-5.1-Codex
+**Known Issues:** Gemini 3 Pro Preview may have format incompatibility
+
+See [Issue #1021](https://github.com/musistudio/claude-code-router/issues/1021) for more details.
+
 #### Custom Router
 
 For more advanced routing logic, you can specify a custom router script via the `CUSTOM_ROUTER_PATH` in your `config.json`. This allows you to implement complex routing rules beyond the default scenarios.
@@ -463,6 +514,63 @@ For routing within subagents, you must specify a particular provider and model b
 Please help me analyze this code snippet for potential optimizations...
 ```
 
+## 🔀 Multiple Instances Support
+
+Run multiple CCR instances simultaneously with different configurations! Perfect for:
+- Different projects requiring different model setups
+- Testing configurations without affecting your main setup
+- Separating work contexts (e.g., personal vs work)
+
+### Usage
+
+**Start a custom instance:**
+```bash
+ccr start --config /path/to/custom-config.json
+```
+
+The port will be auto-allocated if the specified port is busy. Each instance runs independently with its own PID and configuration.
+
+**Use a specific instance:**
+```bash
+ccr code --config /path/to/custom-config.json "your task"
+```
+
+**Check all running instances:**
+```bash
+ccr status                              # Show all instances
+ccr status --config /path/to/config.json  # Show specific instance
+```
+
+**Stop a specific instance:**
+```bash
+ccr stop --config /path/to/custom-config.json
+```
+
+### Example Scenario
+
+```bash
+# Default instance (production setup)
+ccr start
+# -> Runs on port 3456
+
+# Project-specific instance with different models
+ccr start --config ~/project-a/ccr-config.json
+# -> Auto-allocates port 3457
+
+# Another project with local models
+ccr start --config ~/project-b/ccr-config.json
+# -> Auto-allocates port 3458
+
+# View all running instances
+ccr status
+# Shows: Default + 2 custom instances
+
+# Use specific instance
+ccr code --config ~/project-a/ccr-config.json "implement feature X"
+```
+
+See `config.multiinstance.example.json` for a complete example configuration.
+
 ## Status Line (Beta)
 To better monitor the status of claude-code-router at runtime, version v1.0.40 includes a built-in statusline tool, which you can enable in the UI.
 ![statusline-config.png](/blog/images/statusline-config.png)

config.copilot.example.json

@@ -0,0 +1,55 @@
+{
+  "// GITHUB COPILOT API INTEGRATION": "Example configuration for copilot-api",
+  "// Prerequisites": "1. Install copilot-api: https://github.com/ericc-ch/copilot-api",
+  "//               ": "2. Start copilot-api server (default port 4141)",
+  "//               ": "3. Use this config with CCR",
+
+  "LOG": true,
+  "PORT": 3456,
+  "APIKEY": "your-api-key",
+
+  "Providers": [
+    {
+      "name": "copilot",
+      "api_base_url": "http://localhost:4141/v1/messages",
+      "api_key": "dummy-key-for-local-copilot",
+      "models": [
+        "claude-sonnet-4.5",
+        "gpt-5.1",
+        "gpt-5.1-codex"
+      ],
+      "// IMPORTANT": "Use anthropic transformer for proper message format",
+      "transformer": {
+        "use": ["anthropic"]
+      }
+    },
+    {
+      "// Fallback provider": "In case copilot-api is down",
+      "name": "gemini",
+      "api_base_url": "https://generativelanguage.googleapis.com/v1beta/models/",
+      "api_key": "YOUR_GEMINI_API_KEY",
+      "models": [
+        "gemini-2.5-flash",
+        "gemini-2.5-pro"
+      ],
+      "transformer": {
+        "use": ["gemini"]
+      }
+    }
+  ],
+
+  "Router": {
+    "default": "copilot,claude-sonnet-4.5",
+    "background": "copilot,gpt-5.1",
+    "think": "copilot,gpt-5.1",
+    "longContext": "gemini,gemini-2.5-pro",
+    "webSearch": "gemini,gemini-2.5-flash"
+  },
+
+  "// NOTES": {
+    "working_models": "Claude Sonnet 4.5, GPT-5.1, GPT-5.1-Codex work well",
+    "known_issues": "Gemini 3 Pro Preview may have format incompatibility",
+    "solution": "Use anthropic transformer for all Copilot models",
+    "reference": "https://github.com/musistudio/claude-code-router/issues/1021"
+  }
+}

config.example.json

@@ -0,0 +1,134 @@
+{
+  "// BASIC CONFIGURATION": "Claude Code Router - Example Configuration",
+  "// Copy this file to ~/.claude-code-router/config.json and customize it": "",
+
+  "LOG": true,
+  "LOG_LEVEL": "info",
+  "PORT": 3456,
+  "HOST": "127.0.0.1",
+  "APIKEY": "your-api-key-here",
+  "API_TIMEOUT_MS": 600000,
+
+  "// PROVIDERS CONFIGURATION": "Configure your LLM providers below",
+  "Providers": [
+    {
+      "// OpenRouter": "Route to multiple models through OpenRouter",
+      "name": "openrouter",
+      "api_base_url": "https://openrouter.ai/api/v1/chat/completions",
+      "api_key": "sk-or-v1-YOUR_KEY_HERE",
+      "models": [
+        "anthropic/claude-sonnet-4",
+        "anthropic/claude-3.5-sonnet",
+        "google/gemini-2.5-pro-preview",
+        "deepseek/deepseek-chat-v3-0324"
+      ],
+      "transformer": {
+        "use": ["openrouter"],
+        "// Model-specific transformers": "Add anthropic transformer for Claude models",
+        "anthropic/claude-sonnet-4": {
+          "use": ["anthropic", "openrouter"]
+        },
+        "anthropic/claude-3.5-sonnet": {
+          "use": ["anthropic", "openrouter"]
+        }
+      }
+    },
+    {
+      "// DeepSeek": "Direct DeepSeek API",
+      "name": "deepseek",
+      "api_base_url": "https://api.deepseek.com/chat/completions",
+      "api_key": "sk-YOUR_KEY_HERE",
+      "models": [
+        "deepseek-chat",
+        "deepseek-reasoner"
+      ],
+      "transformer": {
+        "use": ["deepseek"],
+        "deepseek-chat": {
+          "use": ["tooluse"]
+        }
+      }
+    },
+    {
+      "// Google Gemini": "Direct Gemini API",
+      "name": "gemini",
+      "api_base_url": "https://generativelanguage.googleapis.com/v1beta/models/",
+      "api_key": "YOUR_GEMINI_API_KEY",
+      "models": [
+        "gemini-2.5-flash",
+        "gemini-2.5-pro"
+      ],
+      "transformer": {
+        "use": ["gemini"]
+      }
+    },
+    {
+      "// Ollama": "Local models via Ollama",
+      "name": "ollama",
+      "api_base_url": "http://localhost:11434/v1/chat/completions",
+      "api_key": "ollama",
+      "models": [
+        "qwen2.5-coder:latest",
+        "deepseek-coder-v2:latest"
+      ]
+    },
+    {
+      "// GitHub Copilot API": "Use copilot-api for multiple models",
+      "// See": "https://github.com/ericc-ch/copilot-api",
+      "name": "copilot",
+      "api_base_url": "http://localhost:4141/v1/messages",
+      "api_key": "dummy-key-for-local",
+      "models": [
+        "claude-sonnet-4.5",
+        "gpt-5.1",
+        "gemini-3-pro-preview"
+      ],
+      "transformer": {
+        "use": ["anthropic"]
+      }
+    },
+    {
+      "// Groq": "Fast inference with Groq",
+      "name": "groq",
+      "api_base_url": "https://api.groq.com/openai/v1/chat/completions",
+      "api_key": "gsk-YOUR_KEY_HERE",
+      "models": [
+        "llama-3.3-70b-versatile",
+        "deepseek-r1-distill-llama-70b"
+      ],
+      "transformer": {
+        "use": ["groq"]
+      }
+    }
+  ],
+
+  "// ROUTER CONFIGURATION": "Define which models to use for different scenarios",
+  "Router": {
+    "default": "openrouter,anthropic/claude-sonnet-4",
+    "background": "gemini,gemini-2.5-flash",
+    "think": "deepseek,deepseek-reasoner",
+    "longContext": "gemini,gemini-2.5-pro",
+    "longContextThreshold": 60000,
+    "webSearch": "openrouter,google/gemini-2.5-pro-preview",
+    "image": "gemini,gemini-2.5-flash"
+  },
+
+  "// IMAGE AGENT": "Force image agent for models without tool calling support",
+  "forceUseImageAgent": false,
+
+  "// STATUS LINE": "Custom status line configuration",
+  "StatusLine": {
+    "enabled": true,
+    "format": "[{sessionId}] {model} | Tokens: {tokens}"
+  },
+
+  "// ADVANCED": "Advanced transformer configurations",
+  "transformers": [
+    {
+      "path": "~/.claude-code-router/plugins/custom-transformer.js",
+      "options": {
+        "customOption": "value"
+      }
+    }
+  ]
+}

config.multiinstance.example.json

@@ -0,0 +1,55 @@
+{
+  "// MULTI-INSTANCE EXAMPLE": "Run multiple CCR instances simultaneously",
+  "// Use Case": "Different models for different projects or workflows",
+  "// This is a sample custom config - save it somewhere like ~/my-project/ccr-config.json",
+
+  "LOG": true,
+  "PORT": 3457,
+  "APIKEY": "custom-instance-key",
+
+  "// CUSTOM INSTANCE SETUP": "This instance uses OpenRouter for main tasks",
+  "Providers": [
+    {
+      "name": "openrouter",
+      "api_base_url": "https://openrouter.ai/api/v1/chat/completions",
+      "api_key": "sk-or-v1-YOUR_KEY",
+      "models": [
+        "anthropic/claude-sonnet-4",
+        "google/gemini-2.5-pro-preview"
+      ],
+      "transformer": {
+        "use": ["openrouter"],
+        "anthropic/claude-sonnet-4": {
+          "use": ["anthropic", "openrouter"]
+        }
+      }
+    },
+    {
+      "name": "ollama",
+      "api_base_url": "http://localhost:11434/v1/chat/completions",
+      "api_key": "ollama",
+      "models": ["qwen2.5-coder:latest"]
+    }
+  ],
+
+  "Router": {
+    "default": "openrouter,anthropic/claude-sonnet-4",
+    "background": "ollama,qwen2.5-coder:latest",
+    "think": "openrouter,google/gemini-2.5-pro-preview"
+  },
+
+  "// USAGE": {
+    "start_instance": "ccr start --config ~/my-project/ccr-config.json",
+    "use_instance": "ccr code --config ~/my-project/ccr-config.json 'your task'",
+    "check_status": "ccr status --config ~/my-project/ccr-config.json",
+    "stop_instance": "ccr stop --config ~/my-project/ccr-config.json",
+    "view_all": "ccr status  # Shows all running instances"
+  },
+
+  "// NOTES": {
+    "port_allocation": "Ports are auto-allocated if specified port is in use",
+    "default_instance": "Default config at ~/.claude-code-router/config.json runs independently",
+    "simultaneous": "You can run unlimited instances with different configs",
+    "isolation": "Each instance has its own PID, port, and configuration"
+  }
+}