documentation of endpoints with i/o structure (Run ID: codestoryai_sidecar_issue_2081_e29258e0)

docs/lsp_configuration.md

@@ -0,0 +1,129 @@
+# Sidecar LSP Configuration Guide
+
+## Overview
+The sidecar service supports flexible Language Server Protocol (LSP) configuration through JSON configuration files. This allows you to:
+- Configure language servers for different programming languages
+- Set custom initialization options
+- Define server-specific settings
+- Manage server lifecycle and capabilities
+
+## Configuration File
+Place your configuration in `lsp_config.json`:
+
+```json
+{
+  "language_servers": {
+    "rust": {
+      "command": "rust-analyzer",
+      "args": [],
+      "initialization_options": {
+        "checkOnSave": true,
+        "procMacro": true
+      },
+      "root_markers": ["Cargo.toml"],
+      "capabilities": [
+        "completions",
+        "diagnostics",
+        "formatting",
+        "references",
+        "definition"
+      ]
+    },
+    "typescript": {
+      "command": "typescript-language-server",
+      "args": ["--stdio"],
+      "initialization_options": {
+        "preferences": {
+          "importModuleSpecifierPreference": "relative"
+        }
+      },
+      "root_markers": ["package.json", "tsconfig.json"],
+      "capabilities": [
+        "completions",
+        "diagnostics",
+        "formatting",
+        "references"
+      ]
+    }
+  },
+  "global_settings": {
+    "workspace_folders": ["src", "tests"],
+    "sync_kind": "full",
+    "completion_trigger_characters": [".", ":", ">"],
+    "signature_trigger_characters": ["(", ","]
+  }
+}
+```
+
+## Configuration Options
+
+### Language Server Configuration
+Configure individual language servers:
+- `command`: Executable name or path
+- `args`: Command line arguments
+- `initialization_options`: LSP initialization parameters
+- `root_markers`: Files indicating project root
+- `capabilities`: Supported LSP features
+
+### Global Settings
+Control LSP behavior across all servers:
+- `workspace_folders`: Default workspace directories
+- `sync_kind`: Document sync type (none/full/incremental)
+- `completion_trigger_characters`: Characters triggering completion
+- `signature_trigger_characters`: Characters triggering signature help
+
+## Supported Languages
+
+### Built-in Support
+- Rust (rust-analyzer)
+- TypeScript/JavaScript (typescript-language-server)
+- Python (pyright)
+- Go (gopls)
+- Java (jdtls)
+- C/C++ (clangd)
+- HTML/CSS (vscode-html-language-server)
+- JSON (vscode-json-language-server)
+- YAML (yaml-language-server)
+- PHP (intelephense)
+
+### Custom Server Configuration
+Example adding a custom language server:
+```json
+{
+  "language_servers": {
+    "custom_lang": {
+      "command": "/path/to/custom-ls",
+      "args": ["--custom-arg"],
+      "initialization_options": {
+        "customSetting": true
+      },
+      "root_markers": ["custom.config"],
+      "capabilities": ["completions", "diagnostics"]
+    }
+  }
+}
+```
+
+## Usage Example
+
+1. Create configuration file:
+```bash
+echo '{
+  "language_servers": {
+    "rust": {
+      "command": "rust-analyzer",
+      "initialization_options": {
+        "checkOnSave": true
+      }
+    }
+  }
+}' > lsp_config.json
+```
+
+2. Load configuration:
+```rust
+let state = LspState::new().await?;
+state.load_configuration(Path::new("lsp_config.json")).await?;
+```
+
+The configuration will be applied automatically to all LSP operations.

docs/model_configuration.md

@@ -0,0 +1,123 @@
+# Sidecar Model Configuration Guide
+
+## Overview
+The sidecar service supports flexible model configuration through JSON configuration files. This allows you to:
+- Enable/disable specific providers or models
+- Override model parameters
+- Configure custom endpoints
+- Set provider-specific settings
+
+## Configuration File
+Place your configuration in `models_config.json`:
+
+```json
+{
+  "config_path": "/path/to/config",
+  "model_overrides": {
+    "gpt-4": {
+      "config": {
+        "temperature": 0.7,
+        "max_tokens": 4096,
+        "top_p": 1.0,
+        "frequency_penalty": 0.0,
+        "presence_penalty": 0.0
+      },
+      "enabled": true,
+      "endpoint": "https://custom-endpoint/v1"
+    },
+    "claude-3-opus": {
+      "config": {
+        "temperature": 0.8,
+        "max_tokens": 8192
+      },
+      "enabled": true
+    }
+  },
+  "enabled_providers": ["OpenAI", "Anthropic", "TogetherAI"],
+  "provider_endpoints": {
+    "OpenAI": "https://api.openai.com/v1",
+    "Anthropic": "https://api.anthropic.com/v1"
+  }
+}
+```
+
+## Configuration Options
+
+### Model Overrides
+Override settings for specific models:
+- `config`: Model-specific parameters
+  - `temperature`: Sampling temperature (0.0-1.0)
+  - `max_tokens`: Maximum tokens to generate
+  - `top_p`: Nucleus sampling parameter
+  - `frequency_penalty`: Frequency penalty for token selection
+  - `presence_penalty`: Presence penalty for token selection
+- `enabled`: Enable/disable specific model
+- `endpoint`: Custom endpoint for this model
+
+### Provider Configuration
+Control provider availability:
+- `enabled_providers`: List of enabled providers (omit to enable all)
+- `provider_endpoints`: Custom endpoints for providers
+
+## Supported Providers & Models
+
+### OpenAI
+- gpt-4-32k
+- gpt-4-preview
+- gpt-4
+- gpt-3.5-turbo-16k
+- gpt-3.5-turbo
+
+### Anthropic
+- claude-3-opus
+- claude-3-sonnet
+- claude-3-haiku
+
+### Together AI
+- codellama-70b-instruct
+- codellama-34b-instruct
+- codellama-13b-instruct
+- llama2-70b
+- llama2-13b
+
+### Google
+- gemini-pro
+- gemini-ultra
+
+### Cohere
+- command-r
+- command
+
+### Mistral
+- mistral-large
+- mistral-medium
+- mistral-small
+
+### Meta
+- llama3-70b
+- llama3-13b
+
+## Usage Example
+
+1. Create configuration file:
+```bash
+echo '{
+  "enabled_providers": ["OpenAI", "Anthropic"],
+  "model_overrides": {
+    "gpt-4": {
+      "config": {
+        "temperature": 0.5
+      },
+      "enabled": true
+    }
+  }
+}' > models_config.json
+```
+
+2. Load configuration:
+```rust
+let state = ModelState::new().await?;
+state.load_configuration(Path::new("models_config.json")).await?;
+```
+
+The configuration will be applied automatically to all model operations.

examples/lsp_config.json

@@ -0,0 +1,74 @@
+{
+    "language_servers": {
+        "rust": {
+            "command": "rust-analyzer",
+            "args": [],
+            "initialization_options": {
+                "checkOnSave": true,
+                "procMacro": true,
+                "diagnostics": {
+                    "enable": true,
+                    "warningsAsHint": []
+                }
+            },
+            "root_markers": ["Cargo.toml"],
+            "capabilities": [
+                "completions",
+                "diagnostics",
+                "formatting",
+                "references",
+                "definition"
+            ]
+        },
+        "typescript": {
+            "command": "typescript-language-server",
+            "args": ["--stdio"],
+            "initialization_options": {
+                "preferences": {
+                    "importModuleSpecifierPreference": "relative"
+                },
+                "typescript": {
+                    "suggest": {
+                        "completeFunctionCalls": true
+                    }
+                }
+            },
+            "root_markers": ["package.json", "tsconfig.json"],
+            "capabilities": [
+                "completions",
+                "diagnostics",
+                "formatting",
+                "references"
+            ]
+        },
+        "python": {
+            "command": "pyright-langserver",
+            "args": ["--stdio"],
+            "initialization_options": {
+                "python": {
+                    "analysis": {
+                        "typeCheckingMode": "basic",
+                        "autoSearchPaths": true
+                    }
+                }
+            },
+            "root_markers": ["pyproject.toml", "setup.py"],
+            "capabilities": [
+                "completions",
+                "diagnostics",
+                "formatting",
+                "references"
+            ]
+        }
+    },
+    "global_settings": {
+        "workspace_folders": ["src", "tests"],
+        "sync_kind": "full",
+        "completion_trigger_characters": [".", ":", ">"],
+        "signature_trigger_characters": ["(", ","],
+        "hover_trigger_characters": [".", ":"],
+        "code_action_trigger_characters": ["."],
+        "format_on_save": true,
+        "max_completion_items": 100
+    }
+}

examples/models_config.json

@@ -0,0 +1,48 @@
+{
+    "enabled_providers": [
+        "OpenAI",
+        "Anthropic",
+        "TogetherAI",
+        "Google",
+        "Cohere",
+        "Mistral",
+        "Meta"
+    ],
+    "model_overrides": {
+        "gpt-4": {
+            "config": {
+                "temperature": 0.7,
+                "max_tokens": 4096,
+                "top_p": 1.0,
+                "frequency_penalty": 0.0,
+                "presence_penalty": 0.0
+            },
+            "enabled": true
+        },
+        "claude-3-opus": {
+            "config": {
+                "temperature": 0.8,
+                "max_tokens": 8192,
+                "top_p": 1.0
+            },
+            "enabled": true
+        },
+        "codellama-70b-instruct": {
+            "config": {
+                "temperature": 0.5,
+                "max_tokens": 4096,
+                "top_p": 0.9
+            },
+            "enabled": true
+        }
+    },
+    "provider_endpoints": {
+        "OpenAI": "https://api.openai.com/v1",
+        "Anthropic": "https://api.anthropic.com/v1",
+        "TogetherAI": "https://api.together.xyz/v1",
+        "Google": "https://generativelanguage.googleapis.com/v1",
+        "Cohere": "https://api.cohere.ai/v1",
+        "Mistral": "https://api.mistral.ai/v1",
+        "Meta": "https://llama.meta.ai/v1"
+    }
+}

fail.txt

@@ -0,0 +1,2 @@
+No changes were made by the agent.
+run_id: codestoryai_sidecar_issue_2081_e29258e0