{
  "$schema": "https://modelcontextprotocol.io/schema/mcp-manifest/v1.json",
  "name": "azara-mcp-server",
  "displayName": "Azara AI Interview Copilot MCP Server",
  "description": "Native Model Context Protocol (MCP) server for Claude, ChatGPT, and autonomous agents to run live interview simulations, analyze technical assessments, and format STAR behavioral answers. Tool discovery is public; tool execution runs a real model but requires an authenticated Azara session.",
  "version": "1.0.0",
  "homepage": "https://azaraai.app/developers",
  "repository": {
    "type": "git",
    "url": "https://github.com/Kryptopacy/Azara"
  },
  "transports": [
    {
      "type": "sse",
      "url": "https://azaraai.app/api/mcp/sse"
    },
    {
      "type": "http",
      "url": "https://azaraai.app/api/mcp"
    }
  ],
  "capabilities": {
    "tools": {
      "listChanged": false
    },
    "resources": {
      "subscribe": false,
      "listChanged": false
    },
    "prompts": {
      "listChanged": false
    }
  },
  "tools": [
    {
      "name": "solve_assessment_question",
      "description": "Analyze and solve live coding questions, quantitative math problems, MCQs, and written technical assessment prompts with verified explanations.",
      "inputSchema": {
        "type": "object",
        "properties": {
          "prompt": {
            "type": "string",
            "description": "The assessment question, coding problem, or case study prompt."
          },
          "context": {
            "type": "string",
            "description": "Optional domain context, programming language, or candidate background."
          },
          "question_type": {
            "type": "string",
            "enum": ["code", "mcq", "analytical", "quant_math", "rlhf", "practical"],
            "description": "Type of assessment question being solved."
          }
        },
        "required": ["prompt"]
      }
    },
    {
      "name": "format_interview_answer",
      "description": "Format concise, punchy spoken bullet points following the STAR (Situation, Task, Action, Result) method for behavioral and executive interview questions.",
      "inputSchema": {
        "type": "object",
        "properties": {
          "question": {
            "type": "string",
            "description": "The behavioral or situational interview question."
          },
          "key_points": {
            "type": "array",
            "items": { "type": "string" },
            "description": "Key achievements, technologies, or metrics to integrate into the STAR framework."
          },
          "role_level": {
            "type": "string",
            "enum": ["entry", "mid", "senior", "lead", "executive"],
            "description": "Target seniority level for voice calibration."
          }
        },
        "required": ["question"]
      }
    },
    {
      "name": "generate_mock_interview_scenario",
      "description": "Generate a realistic, adaptive mock interview session for candidate quality evaluation.",
      "inputSchema": {
        "type": "object",
        "properties": {
          "job_title": {
            "type": "string",
            "description": "Target job title (e.g. Senior Machine Learning Engineer, VP of Product)."
          },
          "difficulty": {
            "type": "string",
            "enum": ["standard", "hard", "faang_bar_raiser", "stress_test"],
            "description": "Target interview difficulty setting."
          }
        },
        "required": ["job_title"]
      }
    }
  ]
}