{
  "schemaVersion": "1.0",
  "item": {
    "slug": "mcp",
    "name": "mcp-builder",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/Veeramanikandanr48/mcp",
    "canonicalUrl": "https://clawhub.ai/Veeramanikandanr48/mcp",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/mcp",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=mcp",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "SKILL.md"
    ],
    "primaryDoc": "SKILL.md",
    "quickSetup": [
      "Download the package from Yavira.",
      "Extract the archive and review SKILL.md first.",
      "Import or place the package into your OpenClaw setup."
    ],
    "agentAssist": {
      "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.",
      "steps": [
        "Download the package from Yavira.",
        "Extract it into a folder your agent can access.",
        "Paste one of the prompts below and point your agent at the extracted folder."
      ],
      "prompts": [
        {
          "label": "New install",
          "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete."
        },
        {
          "label": "Upgrade existing",
          "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run."
        }
      ]
    },
    "sourceHealth": {
      "source": "tencent",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-05-07T17:22:31.273Z",
      "expiresAt": "2026-05-14T17:22:31.273Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=afrexai-annual-report",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=afrexai-annual-report",
        "contentDisposition": "attachment; filename=\"afrexai-annual-report-1.0.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null
      },
      "scope": "source",
      "summary": "Source download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this source.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/mcp"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    },
    "downloadPageUrl": "https://openagent3.xyz/downloads/mcp",
    "agentPageUrl": "https://openagent3.xyz/skills/mcp/agent",
    "manifestUrl": "https://openagent3.xyz/skills/mcp/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/mcp/agent.md"
  },
  "agentAssist": {
    "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.",
    "steps": [
      "Download the package from Yavira.",
      "Extract it into a folder your agent can access.",
      "Paste one of the prompts below and point your agent at the extracted folder."
    ],
    "prompts": [
      {
        "label": "New install",
        "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete."
      },
      {
        "label": "Upgrade existing",
        "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run."
      }
    ]
  },
  "documentation": {
    "source": "clawhub",
    "primaryDoc": "SKILL.md",
    "sections": [
      {
        "title": "Overview",
        "body": "To create high-quality MCP (Model Context Protocol) servers that enable LLMs to effectively interact with external services, use this skill. An MCP server provides tools that allow LLMs to access external services and APIs. The quality of an MCP server is measured by how well it enables LLMs to accomplish real-world tasks using the tools provided."
      },
      {
        "title": "🚀 High-Level Workflow",
        "body": "Creating a high-quality MCP server involves four main phases:"
      },
      {
        "title": "Phase 1: Deep Research and Planning",
        "body": "1.1 Understand Agent-Centric Design Principles\n\nBefore diving into implementation, understand how to design tools for AI agents by reviewing these principles:\n\nBuild for Workflows, Not Just API Endpoints:\n\nDon't simply wrap existing API endpoints - build thoughtful, high-impact workflow tools\nConsolidate related operations (e.g., schedule_event that both checks availability and creates event)\nFocus on tools that enable complete tasks, not just individual API calls\nConsider what workflows agents actually need to accomplish\n\nOptimize for Limited Context:\n\nAgents have constrained context windows - make every token count\nReturn high-signal information, not exhaustive data dumps\nProvide \"concise\" vs \"detailed\" response format options\nDefault to human-readable identifiers over technical codes (names over IDs)\nConsider the agent's context budget as a scarce resource\n\nDesign Actionable Error Messages:\n\nError messages should guide agents toward correct usage patterns\nSuggest specific next steps: \"Try using filter='active_only' to reduce results\"\nMake errors educational, not just diagnostic\nHelp agents learn proper tool usage through clear feedback\n\nFollow Natural Task Subdivisions:\n\nTool names should reflect how humans think about tasks\nGroup related tools with consistent prefixes for discoverability\nDesign tools around natural workflows, not just API structure\n\nUse Evaluation-Driven Development:\n\nCreate realistic evaluation scenarios early\nLet agent feedback drive tool improvements\nPrototype quickly and iterate based on actual agent performance\n\n1.3 Study MCP Protocol Documentation\n\nFetch the latest MCP protocol documentation:\n\nUse WebFetch to load: https://modelcontextprotocol.io/llms-full.txt\n\nThis comprehensive document contains the complete MCP specification and guidelines.\n\n1.4 Study Framework Documentation\n\nLoad and read the following reference files:\n\nMCP Best Practices: 📋 View Best Practices - Core guidelines for all MCP servers\n\nFor Python implementations, also load:\n\nPython SDK Documentation: Use WebFetch to load https://raw.githubusercontent.com/modelcontextprotocol/python-sdk/main/README.md\n🐍 Python Implementation Guide - Python-specific best practices and examples\n\nFor Node/TypeScript implementations, also load:\n\nTypeScript SDK Documentation: Use WebFetch to load https://raw.githubusercontent.com/modelcontextprotocol/typescript-sdk/main/README.md\n⚡ TypeScript Implementation Guide - Node/TypeScript-specific best practices and examples\n\n1.5 Exhaustively Study API Documentation\n\nTo integrate a service, read through ALL available API documentation:\n\nOfficial API reference documentation\nAuthentication and authorization requirements\nRate limiting and pagination patterns\nError responses and status codes\nAvailable endpoints and their parameters\nData models and schemas\n\nTo gather comprehensive information, use web search and the WebFetch tool as needed.\n\n1.6 Create a Comprehensive Implementation Plan\n\nBased on your research, create a detailed plan that includes:\n\nTool Selection:\n\nList the most valuable endpoints/operations to implement\nPrioritize tools that enable the most common and important use cases\nConsider which tools work together to enable complex workflows\n\nShared Utilities and Helpers:\n\nIdentify common API request patterns\nPlan pagination helpers\nDesign filtering and formatting utilities\nPlan error handling strategies\n\nInput/Output Design:\n\nDefine input validation models (Pydantic for Python, Zod for TypeScript)\nDesign consistent response formats (e.g., JSON or Markdown), and configurable levels of detail (e.g., Detailed or Concise)\nPlan for large-scale usage (thousands of users/resources)\nImplement character limits and truncation strategies (e.g., 25,000 tokens)\n\nError Handling Strategy:\n\nPlan graceful failure modes\nDesign clear, actionable, LLM-friendly, natural language error messages which prompt further action\nConsider rate limiting and timeout scenarios\nHandle authentication and authorization errors"
      },
      {
        "title": "Phase 2: Implementation",
        "body": "Now that you have a comprehensive plan, begin implementation following language-specific best practices.\n\n2.1 Set Up Project Structure\n\nFor Python:\n\nCreate a single .py file or organize into modules if complex (see 🐍 Python Guide)\nUse the MCP Python SDK for tool registration\nDefine Pydantic models for input validation\n\nFor Node/TypeScript:\n\nCreate proper project structure (see ⚡ TypeScript Guide)\nSet up package.json and tsconfig.json\nUse MCP TypeScript SDK\nDefine Zod schemas for input validation\n\n2.2 Implement Core Infrastructure First\n\nTo begin implementation, create shared utilities before implementing tools:\n\nAPI request helper functions\nError handling utilities\nResponse formatting functions (JSON and Markdown)\nPagination helpers\nAuthentication/token management\n\n2.3 Implement Tools Systematically\n\nFor each tool in the plan:\n\nDefine Input Schema:\n\nUse Pydantic (Python) or Zod (TypeScript) for validation\nInclude proper constraints (min/max length, regex patterns, min/max values, ranges)\nProvide clear, descriptive field descriptions\nInclude diverse examples in field descriptions\n\nWrite Comprehensive Docstrings/Descriptions:\n\nOne-line summary of what the tool does\nDetailed explanation of purpose and functionality\nExplicit parameter types with examples\nComplete return type schema\nUsage examples (when to use, when not to use)\nError handling documentation, which outlines how to proceed given specific errors\n\nImplement Tool Logic:\n\nUse shared utilities to avoid code duplication\nFollow async/await patterns for all I/O\nImplement proper error handling\nSupport multiple response formats (JSON and Markdown)\nRespect pagination parameters\nCheck character limits and truncate appropriately\n\nAdd Tool Annotations:\n\nreadOnlyHint: true (for read-only operations)\ndestructiveHint: false (for non-destructive operations)\nidempotentHint: true (if repeated calls have same effect)\nopenWorldHint: true (if interacting with external systems)\n\n2.4 Follow Language-Specific Best Practices\n\nAt this point, load the appropriate language guide:\n\nFor Python: Load 🐍 Python Implementation Guide and ensure the following:\n\nUsing MCP Python SDK with proper tool registration\nPydantic v2 models with model_config\nType hints throughout\nAsync/await for all I/O operations\nProper imports organization\nModule-level constants (CHARACTER_LIMIT, API_BASE_URL)\n\nFor Node/TypeScript: Load ⚡ TypeScript Implementation Guide and ensure the following:\n\nUsing server.registerTool properly\nZod schemas with .strict()\nTypeScript strict mode enabled\nNo any types - use proper types\nExplicit Promise<T> return types\nBuild process configured (npm run build)"
      },
      {
        "title": "Phase 3: Review and Refine",
        "body": "After initial implementation:\n\n3.1 Code Quality Review\n\nTo ensure quality, review the code for:\n\nDRY Principle: No duplicated code between tools\nComposability: Shared logic extracted into functions\nConsistency: Similar operations return similar formats\nError Handling: All external calls have error handling\nType Safety: Full type coverage (Python type hints, TypeScript types)\nDocumentation: Every tool has comprehensive docstrings/descriptions\n\n3.2 Test and Build\n\nImportant: MCP servers are long-running processes that wait for requests over stdio/stdin or sse/http. Running them directly in your main process (e.g., python server.py or node dist/index.js) will cause your process to hang indefinitely.\n\nSafe ways to test the server:\n\nUse the evaluation harness (see Phase 4) - recommended approach\nRun the server in tmux to keep it outside your main process\nUse a timeout when testing: timeout 5s python server.py\n\nFor Python:\n\nVerify Python syntax: python -m py_compile your_server.py\nCheck imports work correctly by reviewing the file\nTo manually test: Run server in tmux, then test with evaluation harness in main process\nOr use the evaluation harness directly (it manages the server for stdio transport)\n\nFor Node/TypeScript:\n\nRun npm run build and ensure it completes without errors\nVerify dist/index.js is created\nTo manually test: Run server in tmux, then test with evaluation harness in main process\nOr use the evaluation harness directly (it manages the server for stdio transport)\n\n3.3 Use Quality Checklist\n\nTo verify implementation quality, load the appropriate checklist from the language-specific guide:\n\nPython: see \"Quality Checklist\" in 🐍 Python Guide\nNode/TypeScript: see \"Quality Checklist\" in ⚡ TypeScript Guide"
      },
      {
        "title": "Phase 4: Create Evaluations",
        "body": "After implementing your MCP server, create comprehensive evaluations to test its effectiveness.\n\nLoad ✅ Evaluation Guide for complete evaluation guidelines.\n\n4.1 Understand Evaluation Purpose\n\nEvaluations test whether LLMs can effectively use your MCP server to answer realistic, complex questions.\n\n4.2 Create 10 Evaluation Questions\n\nTo create effective evaluations, follow the process outlined in the evaluation guide:\n\nTool Inspection: List available tools and understand their capabilities\nContent Exploration: Use READ-ONLY operations to explore available data\nQuestion Generation: Create 10 complex, realistic questions\nAnswer Verification: Solve each question yourself to verify answers\n\n4.3 Evaluation Requirements\n\nEach question must be:\n\nIndependent: Not dependent on other questions\nRead-only: Only non-destructive operations required\nComplex: Requiring multiple tool calls and deep exploration\nRealistic: Based on real use cases humans would care about\nVerifiable: Single, clear answer that can be verified by string comparison\nStable: Answer won't change over time\n\n4.4 Output Format\n\nCreate an XML file with this structure:\n\n<evaluation>\n  <qa_pair>\n    <question>Find discussions about AI model launches with animal codenames. One model needed a specific safety designation that uses the format ASL-X. What number X was being determined for the model named after a spotted wild cat?</question>\n    <answer>3</answer>\n  </qa_pair>\n<!-- More qa_pairs... -->\n</evaluation>"
      },
      {
        "title": "📚 Documentation Library",
        "body": "Load these resources as needed during development:"
      },
      {
        "title": "Core MCP Documentation (Load First)",
        "body": "MCP Protocol: Fetch from https://modelcontextprotocol.io/llms-full.txt - Complete MCP specification\n📋 MCP Best Practices - Universal MCP guidelines including:\n\nServer and tool naming conventions\nResponse format guidelines (JSON vs Markdown)\nPagination best practices\nCharacter limits and truncation strategies\nTool development guidelines\nSecurity and error handling standards"
      },
      {
        "title": "SDK Documentation (Load During Phase 1/2)",
        "body": "Python SDK: Fetch from https://raw.githubusercontent.com/modelcontextprotocol/python-sdk/main/README.md\nTypeScript SDK: Fetch from https://raw.githubusercontent.com/modelcontextprotocol/typescript-sdk/main/README.md"
      },
      {
        "title": "Language-Specific Implementation Guides (Load During Phase 2)",
        "body": "🐍 Python Implementation Guide - Complete Python/FastMCP guide with:\n\nServer initialization patterns\nPydantic model examples\nTool registration with @mcp.tool\nComplete working examples\nQuality checklist\n\n\n\n⚡ TypeScript Implementation Guide - Complete TypeScript guide with:\n\nProject structure\nZod schema patterns\nTool registration with server.registerTool\nComplete working examples\nQuality checklist"
      },
      {
        "title": "Evaluation Guide (Load During Phase 4)",
        "body": "✅ Evaluation Guide - Complete evaluation creation guide with:\n\nQuestion creation guidelines\nAnswer verification strategies\nXML format specifications\nExample questions and answers\nRunning an evaluation with the provided scripts"
      }
    ],
    "body": "MCP Server Development Guide\nOverview\n\nTo create high-quality MCP (Model Context Protocol) servers that enable LLMs to effectively interact with external services, use this skill. An MCP server provides tools that allow LLMs to access external services and APIs. The quality of an MCP server is measured by how well it enables LLMs to accomplish real-world tasks using the tools provided.\n\nProcess\n🚀 High-Level Workflow\n\nCreating a high-quality MCP server involves four main phases:\n\nPhase 1: Deep Research and Planning\n1.1 Understand Agent-Centric Design Principles\n\nBefore diving into implementation, understand how to design tools for AI agents by reviewing these principles:\n\nBuild for Workflows, Not Just API Endpoints:\n\nDon't simply wrap existing API endpoints - build thoughtful, high-impact workflow tools\nConsolidate related operations (e.g., schedule_event that both checks availability and creates event)\nFocus on tools that enable complete tasks, not just individual API calls\nConsider what workflows agents actually need to accomplish\n\nOptimize for Limited Context:\n\nAgents have constrained context windows - make every token count\nReturn high-signal information, not exhaustive data dumps\nProvide \"concise\" vs \"detailed\" response format options\nDefault to human-readable identifiers over technical codes (names over IDs)\nConsider the agent's context budget as a scarce resource\n\nDesign Actionable Error Messages:\n\nError messages should guide agents toward correct usage patterns\nSuggest specific next steps: \"Try using filter='active_only' to reduce results\"\nMake errors educational, not just diagnostic\nHelp agents learn proper tool usage through clear feedback\n\nFollow Natural Task Subdivisions:\n\nTool names should reflect how humans think about tasks\nGroup related tools with consistent prefixes for discoverability\nDesign tools around natural workflows, not just API structure\n\nUse Evaluation-Driven Development:\n\nCreate realistic evaluation scenarios early\nLet agent feedback drive tool improvements\nPrototype quickly and iterate based on actual agent performance\n1.3 Study MCP Protocol Documentation\n\nFetch the latest MCP protocol documentation:\n\nUse WebFetch to load: https://modelcontextprotocol.io/llms-full.txt\n\nThis comprehensive document contains the complete MCP specification and guidelines.\n\n1.4 Study Framework Documentation\n\nLoad and read the following reference files:\n\nMCP Best Practices: 📋 View Best Practices - Core guidelines for all MCP servers\n\nFor Python implementations, also load:\n\nPython SDK Documentation: Use WebFetch to load https://raw.githubusercontent.com/modelcontextprotocol/python-sdk/main/README.md\n🐍 Python Implementation Guide - Python-specific best practices and examples\n\nFor Node/TypeScript implementations, also load:\n\nTypeScript SDK Documentation: Use WebFetch to load https://raw.githubusercontent.com/modelcontextprotocol/typescript-sdk/main/README.md\n⚡ TypeScript Implementation Guide - Node/TypeScript-specific best practices and examples\n1.5 Exhaustively Study API Documentation\n\nTo integrate a service, read through ALL available API documentation:\n\nOfficial API reference documentation\nAuthentication and authorization requirements\nRate limiting and pagination patterns\nError responses and status codes\nAvailable endpoints and their parameters\nData models and schemas\n\nTo gather comprehensive information, use web search and the WebFetch tool as needed.\n\n1.6 Create a Comprehensive Implementation Plan\n\nBased on your research, create a detailed plan that includes:\n\nTool Selection:\n\nList the most valuable endpoints/operations to implement\nPrioritize tools that enable the most common and important use cases\nConsider which tools work together to enable complex workflows\n\nShared Utilities and Helpers:\n\nIdentify common API request patterns\nPlan pagination helpers\nDesign filtering and formatting utilities\nPlan error handling strategies\n\nInput/Output Design:\n\nDefine input validation models (Pydantic for Python, Zod for TypeScript)\nDesign consistent response formats (e.g., JSON or Markdown), and configurable levels of detail (e.g., Detailed or Concise)\nPlan for large-scale usage (thousands of users/resources)\nImplement character limits and truncation strategies (e.g., 25,000 tokens)\n\nError Handling Strategy:\n\nPlan graceful failure modes\nDesign clear, actionable, LLM-friendly, natural language error messages which prompt further action\nConsider rate limiting and timeout scenarios\nHandle authentication and authorization errors\nPhase 2: Implementation\n\nNow that you have a comprehensive plan, begin implementation following language-specific best practices.\n\n2.1 Set Up Project Structure\n\nFor Python:\n\nCreate a single .py file or organize into modules if complex (see 🐍 Python Guide)\nUse the MCP Python SDK for tool registration\nDefine Pydantic models for input validation\n\nFor Node/TypeScript:\n\nCreate proper project structure (see ⚡ TypeScript Guide)\nSet up package.json and tsconfig.json\nUse MCP TypeScript SDK\nDefine Zod schemas for input validation\n2.2 Implement Core Infrastructure First\n\nTo begin implementation, create shared utilities before implementing tools:\n\nAPI request helper functions\nError handling utilities\nResponse formatting functions (JSON and Markdown)\nPagination helpers\nAuthentication/token management\n2.3 Implement Tools Systematically\n\nFor each tool in the plan:\n\nDefine Input Schema:\n\nUse Pydantic (Python) or Zod (TypeScript) for validation\nInclude proper constraints (min/max length, regex patterns, min/max values, ranges)\nProvide clear, descriptive field descriptions\nInclude diverse examples in field descriptions\n\nWrite Comprehensive Docstrings/Descriptions:\n\nOne-line summary of what the tool does\nDetailed explanation of purpose and functionality\nExplicit parameter types with examples\nComplete return type schema\nUsage examples (when to use, when not to use)\nError handling documentation, which outlines how to proceed given specific errors\n\nImplement Tool Logic:\n\nUse shared utilities to avoid code duplication\nFollow async/await patterns for all I/O\nImplement proper error handling\nSupport multiple response formats (JSON and Markdown)\nRespect pagination parameters\nCheck character limits and truncate appropriately\n\nAdd Tool Annotations:\n\nreadOnlyHint: true (for read-only operations)\ndestructiveHint: false (for non-destructive operations)\nidempotentHint: true (if repeated calls have same effect)\nopenWorldHint: true (if interacting with external systems)\n2.4 Follow Language-Specific Best Practices\n\nAt this point, load the appropriate language guide:\n\nFor Python: Load 🐍 Python Implementation Guide and ensure the following:\n\nUsing MCP Python SDK with proper tool registration\nPydantic v2 models with model_config\nType hints throughout\nAsync/await for all I/O operations\nProper imports organization\nModule-level constants (CHARACTER_LIMIT, API_BASE_URL)\n\nFor Node/TypeScript: Load ⚡ TypeScript Implementation Guide and ensure the following:\n\nUsing server.registerTool properly\nZod schemas with .strict()\nTypeScript strict mode enabled\nNo any types - use proper types\nExplicit Promise<T> return types\nBuild process configured (npm run build)\nPhase 3: Review and Refine\n\nAfter initial implementation:\n\n3.1 Code Quality Review\n\nTo ensure quality, review the code for:\n\nDRY Principle: No duplicated code between tools\nComposability: Shared logic extracted into functions\nConsistency: Similar operations return similar formats\nError Handling: All external calls have error handling\nType Safety: Full type coverage (Python type hints, TypeScript types)\nDocumentation: Every tool has comprehensive docstrings/descriptions\n3.2 Test and Build\n\nImportant: MCP servers are long-running processes that wait for requests over stdio/stdin or sse/http. Running them directly in your main process (e.g., python server.py or node dist/index.js) will cause your process to hang indefinitely.\n\nSafe ways to test the server:\n\nUse the evaluation harness (see Phase 4) - recommended approach\nRun the server in tmux to keep it outside your main process\nUse a timeout when testing: timeout 5s python server.py\n\nFor Python:\n\nVerify Python syntax: python -m py_compile your_server.py\nCheck imports work correctly by reviewing the file\nTo manually test: Run server in tmux, then test with evaluation harness in main process\nOr use the evaluation harness directly (it manages the server for stdio transport)\n\nFor Node/TypeScript:\n\nRun npm run build and ensure it completes without errors\nVerify dist/index.js is created\nTo manually test: Run server in tmux, then test with evaluation harness in main process\nOr use the evaluation harness directly (it manages the server for stdio transport)\n3.3 Use Quality Checklist\n\nTo verify implementation quality, load the appropriate checklist from the language-specific guide:\n\nPython: see \"Quality Checklist\" in 🐍 Python Guide\nNode/TypeScript: see \"Quality Checklist\" in ⚡ TypeScript Guide\nPhase 4: Create Evaluations\n\nAfter implementing your MCP server, create comprehensive evaluations to test its effectiveness.\n\nLoad ✅ Evaluation Guide for complete evaluation guidelines.\n\n4.1 Understand Evaluation Purpose\n\nEvaluations test whether LLMs can effectively use your MCP server to answer realistic, complex questions.\n\n4.2 Create 10 Evaluation Questions\n\nTo create effective evaluations, follow the process outlined in the evaluation guide:\n\nTool Inspection: List available tools and understand their capabilities\nContent Exploration: Use READ-ONLY operations to explore available data\nQuestion Generation: Create 10 complex, realistic questions\nAnswer Verification: Solve each question yourself to verify answers\n4.3 Evaluation Requirements\n\nEach question must be:\n\nIndependent: Not dependent on other questions\nRead-only: Only non-destructive operations required\nComplex: Requiring multiple tool calls and deep exploration\nRealistic: Based on real use cases humans would care about\nVerifiable: Single, clear answer that can be verified by string comparison\nStable: Answer won't change over time\n4.4 Output Format\n\nCreate an XML file with this structure:\n\n<evaluation>\n  <qa_pair>\n    <question>Find discussions about AI model launches with animal codenames. One model needed a specific safety designation that uses the format ASL-X. What number X was being determined for the model named after a spotted wild cat?</question>\n    <answer>3</answer>\n  </qa_pair>\n<!-- More qa_pairs... -->\n</evaluation>\n\nReference Files\n📚 Documentation Library\n\nLoad these resources as needed during development:\n\nCore MCP Documentation (Load First)\nMCP Protocol: Fetch from https://modelcontextprotocol.io/llms-full.txt - Complete MCP specification\n📋 MCP Best Practices - Universal MCP guidelines including:\nServer and tool naming conventions\nResponse format guidelines (JSON vs Markdown)\nPagination best practices\nCharacter limits and truncation strategies\nTool development guidelines\nSecurity and error handling standards\nSDK Documentation (Load During Phase 1/2)\nPython SDK: Fetch from https://raw.githubusercontent.com/modelcontextprotocol/python-sdk/main/README.md\nTypeScript SDK: Fetch from https://raw.githubusercontent.com/modelcontextprotocol/typescript-sdk/main/README.md\nLanguage-Specific Implementation Guides (Load During Phase 2)\n\n🐍 Python Implementation Guide - Complete Python/FastMCP guide with:\n\nServer initialization patterns\nPydantic model examples\nTool registration with @mcp.tool\nComplete working examples\nQuality checklist\n\n⚡ TypeScript Implementation Guide - Complete TypeScript guide with:\n\nProject structure\nZod schema patterns\nTool registration with server.registerTool\nComplete working examples\nQuality checklist\nEvaluation Guide (Load During Phase 4)\n✅ Evaluation Guide - Complete evaluation creation guide with:\nQuestion creation guidelines\nAnswer verification strategies\nXML format specifications\nExample questions and answers\nRunning an evaluation with the provided scripts"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/Veeramanikandanr48/mcp",
    "publisherUrl": "https://clawhub.ai/Veeramanikandanr48/mcp",
    "owner": "Veeramanikandanr48",
    "version": "1.0.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/mcp",
    "downloadUrl": "https://openagent3.xyz/downloads/mcp",
    "agentUrl": "https://openagent3.xyz/skills/mcp/agent",
    "manifestUrl": "https://openagent3.xyz/skills/mcp/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/mcp/agent.md"
  }
}