Sequential Questioning MCP Server

# API Reference This document provides a complete reference for all API endpoints in the Sequential Questioning MCP Server. ## MCP Endpoints ### Generate Sequential Question Generates a contextually relevant question based on conversation history. ``` POST /mcp/sequential-questioning ``` #### Request Body ```json { "user_id": "string", "context": "string", "previous_messages": [ { "role": "string", "content": "string" } ] } ``` **Parameters:** - `user_id` (required): Unique identifier for the user - `context` (required): Initial context for the conversation - `previous_messages` (optional): List of previous messages in the conversation - `role`: Either "user" or "assistant" - `content`: The content of the message #### Response ```json { "conversation_id": "string", "session_id": "string", "question": "string", "metadata": { "generation_time": "float", "tokens_used": "integer", "similar_contexts": [ "string" ] } } ``` **Fields:** - `conversation_id`: Unique identifier for the conversation - `session_id`: Unique identifier for the user session - `question`: The generated question - `metadata`: Additional information about the generation process - `generation_time`: Time taken to generate the question in seconds - `tokens_used`: Number of tokens used in the generation process - `similar_contexts`: Similar contexts found in the vector database (if any) #### Error Responses - **400 Bad Request**: Invalid request parameters - **500 Internal Server Error**: Failed to generate question ## Internal Endpoints ### Health Check ``` GET /health ``` #### Response ```json { "status": "ok", "version": "string", "timestamp": "string" } ``` ### Metrics ``` GET /mcp-internal/monitoring/metrics ``` #### Response ```json { "requests": { "total": "integer", "by_endpoint": { "endpoint_path": "integer" } }, "errors": { "total": "integer", "by_type": { "error_type": "integer" } }, "response_times": { "avg": "float", "min": "float", "max": "float", "p95": "float", "p99": "float" } } ``` ### Reset Metrics ``` POST /mcp-internal/monitoring/reset ``` #### Response ```json { "status": "success", "message": "Metrics reset successfully" } ``` ## Using the API ### Authentication Currently, the API does not require authentication. This will be added in future versions. ### Rate Limiting Rate limiting is not implemented in the current version. ### Example Request (cURL) ```bash curl -X POST "http://localhost:8000/mcp/sequential-questioning" \ -H "Content-Type: application/json" \ -d '{ "user_id": "user123", "context": "We are discussing climate change and its effects on agriculture.", "previous_messages": [ { "role": "user", "content": "How does rising temperature affect crop yields?" }, { "role": "assistant", "content": "Rising temperatures can reduce crop yields through heat stress, increased water demand, and changes in pest populations." } ] }' ``` ### Example Response ```json { "conversation_id": "conv_9f8a7b6c5d", "session_id": "sess_1a2b3c4d5e", "question": "What specific crops are most vulnerable to these temperature increases, and are there any regions particularly at risk?", "metadata": { "generation_time": 1.25, "tokens_used": 128, "similar_contexts": [ "Discussion about drought-resistant crops in warming climates" ] } } ```