Skip to content

aiskillstore/marketplace / skills/92bilal26/mcp-builder

mcp-builder

Verified

Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).

skill-report.json

View on GitHub
{
  "schema_version": "2.0",
  "meta": {
    "generated_at": "2026-07-05T15:29:47.815Z",
    "slug": "92bilal26-mcp-builder",
    "source_url": "https://github.com/92Bilal26/TaskPilotAI/tree/main/.claude/skills/mcp-builder",
    "source_ref": "main",
    "model": "codex",
    "analysis_version": "3.0.0",
    "source_type": "community",
    "content_hash": "0f4592dcb53cf2b5d6b7febee6b4152018b565551a1c29e3c612f57b218ab295",
    "tree_hash": "88d0330cb843321d7dee80a53cff04d01bc9328cb96cc603f2ae28bb6a0713c6",
    "provenance": {
      "model_effective": "codex",
      "fallback_chain": [
        {
          "agent": "codex",
          "outcome": "success"
        }
      ],
      "analysis_version": "3.0.0"
    }
  },
  "skill": {
    "name": "mcp-builder",
    "description": "Guide for creating high-quality MCP (Model Context Protocol) servers that connect LLMs to external services through well-designed tools, schemas, transports, and evaluations.",
    "summary": "Build, review, and evaluate Python or TypeScript MCP servers with practical implementation guidance.",
    "icon": "📦",
    "version": "1.0.1",
    "author": "92Bilal26",
    "license": "Complete terms in LICENSE.txt",
    "tags": [
      "mcp",
      "model-context-protocol",
      "typescript",
      "python",
      "llm-tools"
    ],
    "supported_tools": [
      "claude",
      "codex",
      "claude-code"
    ],
    "risk_factors": [
      "network",
      "env_access",
      "external_commands"
    ],
    "category": "coding"
  },
  "security_audit": {
    "risk_level": "high",
    "is_blocked": false,
    "safe_to_publish": false,
    "analysis_status": "ok",
    "summary": "Most static findings are false positives caused by Markdown examples, TypeScript template literals, placeholder URLs, and dummy credential names in reference documentation. No prompt-injection attempt or embedded real secret was found. The remaining concerns are operational: the evaluation harness can launch user-supplied commands, pass environment values, connect to arbitrary MCP URLs, and send tool outputs to Anthropic.",
    "remediation": [
      {
        "issue": "Evaluation harness can execute caller-supplied local commands.",
        "suggestion": "Document that users must only run trusted commands and trusted evaluation files. Consider adding command allowlists or confirmation prompts before launching stdio servers.",
        "severity": "high"
      },
      {
        "issue": "Evaluation runs can send questions, tool schemas, and tool outputs to Anthropic.",
        "suggestion": "Warn users to avoid sensitive production data during evaluations. Add optional redaction hooks for tool output, headers, and report content.",
        "severity": "medium"
      },
      {
        "issue": "Remote URL and header options can send tokens to arbitrary MCP endpoints.",
        "suggestion": "Require users to verify remote MCP URLs before passing Authorization headers. Prefer test credentials with limited scope.",
        "severity": "medium"
      },
      {
        "issue": "Documentation includes placeholder tokens and command examples.",
        "suggestion": "Label every placeholder credential as non-secret sample data and remind users not to paste real tokens into shared files or reports.",
        "severity": "low"
      }
    ],
    "risk_factor_evidence": [
      {
        "factor": "network",
        "evidence": [
          {
            "file": "reference/evaluation.md",
            "line_start": 456,
            "line_end": 456
          },
          {
            "file": "reference/evaluation.md",
            "line_start": 469,
            "line_end": 469
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 413,
            "line_end": 413
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 449,
            "line_end": 449
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 474,
            "line_end": 474
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 480,
            "line_end": 480
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 548,
            "line_end": 548
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 598,
            "line_end": 598
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 642,
            "line_end": 642
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 942,
            "line_end": 942
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 424,
            "line_end": 424
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 668,
            "line_end": 668
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 601,
            "line_end": 601
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 739,
            "line_end": 739
          },
          {
            "file": "reference/python_mcp_server.md",
            "line_start": 193,
            "line_end": 193
          },
          {
            "file": "reference/python_mcp_server.md",
            "line_start": 233,
            "line_end": 233
          },
          {
            "file": "reference/python_mcp_server.md",
            "line_start": 236,
            "line_end": 236
          },
          {
            "file": "reference/python_mcp_server.md",
            "line_start": 382,
            "line_end": 382
          },
          {
            "file": "reference/python_mcp_server.md",
            "line_start": 385,
            "line_end": 385
          },
          {
            "file": "reference/python_mcp_server.md",
            "line_start": 426,
            "line_end": 426
          },
          {
            "file": "reference/python_mcp_server.md",
            "line_start": 220,
            "line_end": 220
          },
          {
            "file": "reference/python_mcp_server.md",
            "line_start": 260,
            "line_end": 260
          },
          {
            "file": "reference/python_mcp_server.md",
            "line_start": 402,
            "line_end": 402
          },
          {
            "file": "reference/python_mcp_server.md",
            "line_start": 353,
            "line_end": 353
          },
          {
            "file": "scripts/evaluation.py",
            "line_start": 315,
            "line_end": 315
          },
          {
            "file": "scripts/evaluation.py",
            "line_start": 318,
            "line_end": 318
          },
          {
            "file": "SKILL.md",
            "line_start": 41,
            "line_end": 41
          },
          {
            "file": "SKILL.md",
            "line_start": 43,
            "line_end": 43
          },
          {
            "file": "SKILL.md",
            "line_start": 61,
            "line_end": 61
          },
          {
            "file": "SKILL.md",
            "line_start": 65,
            "line_end": 65
          },
          {
            "file": "SKILL.md",
            "line_start": 203,
            "line_end": 203
          },
          {
            "file": "SKILL.md",
            "line_start": 212,
            "line_end": 212
          },
          {
            "file": "SKILL.md",
            "line_start": 213,
            "line_end": 213
          }
        ]
      },
      {
        "factor": "env_access",
        "evidence": [
          {
            "file": "reference/evaluation.md",
            "line_start": 567,
            "line_end": 567
          },
          {
            "file": "reference/evaluation.md",
            "line_start": 398,
            "line_end": 398
          },
          {
            "file": "reference/evaluation.md",
            "line_start": 444,
            "line_end": 444
          },
          {
            "file": "reference/evaluation.md",
            "line_start": 557,
            "line_end": 557
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 707,
            "line_end": 707
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 719,
            "line_end": 719
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 737,
            "line_end": 737
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 744,
            "line_end": 744
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 707,
            "line_end": 707
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 719,
            "line_end": 719
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 737,
            "line_end": 737
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 744,
            "line_end": 744
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 707,
            "line_end": 707
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 708,
            "line_end": 708
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 719,
            "line_end": 719
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 720,
            "line_end": 720
          },
          {
            "file": "reference/python_mcp_server.md",
            "line_start": 511,
            "line_end": 511
          },
          {
            "file": "reference/python_mcp_server.md",
            "line_start": 517,
            "line_end": 517
          }
        ]
      },
      {
        "factor": "external_commands",
        "evidence": [
          {
            "file": "reference/mcp_best_practices.md",
            "line_start": 223,
            "line_end": 227
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 39,
            "line_end": 46
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 190,
            "line_end": 221
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 221,
            "line_end": 247
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 247,
            "line_end": 248
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 248,
            "line_end": 250
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 250,
            "line_end": 251
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 251,
            "line_end": 252
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 252,
            "line_end": 274
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 399,
            "line_end": 400
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 400,
            "line_end": 406
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 426,
            "line_end": 432
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 432,
            "line_end": 434
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 451,
            "line_end": 465
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 474,
            "line_end": 480
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 480,
            "line_end": 483
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 516,
            "line_end": 522
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 522,
            "line_end": 524
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 644,
            "line_end": 670
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 670,
            "line_end": 676
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 676,
            "line_end": 690
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 690,
            "line_end": 739
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 739,
            "line_end": 756
          },
          {
            "file": "reference/node_mcp_server.md",
            "line_start": 802,
            "line_end": 809
          },
          {
            "file": "scripts/evaluation.py",
            "line_start": 204,
            "line_end": 204
          },
          {
            "file": "scripts/evaluation.py",
            "line_start": 205,
            "line_end": 205
          },
          {
            "file": "SKILL.md",
            "line_start": 29,
            "line_end": 29
          },
          {
            "file": "SKILL.md",
            "line_start": 41,
            "line_end": 41
          },
          {
            "file": "SKILL.md",
            "line_start": 43,
            "line_end": 43
          },
          {
            "file": "SKILL.md",
            "line_start": 61,
            "line_end": 61
          },
          {
            "file": "SKILL.md",
            "line_start": 65,
            "line_end": 65
          },
          {
            "file": "SKILL.md",
            "line_start": 104,
            "line_end": 104
          },
          {
            "file": "SKILL.md",
            "line_start": 105,
            "line_end": 105
          },
          {
            "file": "SKILL.md",
            "line_start": 120,
            "line_end": 120
          },
          {
            "file": "SKILL.md",
            "line_start": 121,
            "line_end": 121
          },
          {
            "file": "SKILL.md",
            "line_start": 122,
            "line_end": 122
          },
          {
            "file": "SKILL.md",
            "line_start": 123,
            "line_end": 123
          },
          {
            "file": "SKILL.md",
            "line_start": 140,
            "line_end": 140
          },
          {
            "file": "SKILL.md",
            "line_start": 141,
            "line_end": 141
          },
          {
            "file": "SKILL.md",
            "line_start": 144,
            "line_end": 144
          },
          {
            "file": "SKILL.md",
            "line_start": 184,
            "line_end": 192
          },
          {
            "file": "SKILL.md",
            "line_start": 192,
            "line_end": 203
          },
          {
            "file": "SKILL.md",
            "line_start": 203,
            "line_end": 203
          },
          {
            "file": "SKILL.md",
            "line_start": 212,
            "line_end": 213
          },
          {
            "file": "SKILL.md",
            "line_start": 213,
            "line_end": 219
          },
          {
            "file": "SKILL.md",
            "line_start": 219,
            "line_end": 226
          }
        ]
      }
    ],
    "static_findings": [
      {
        "id": "network:reference/evaluation.md:456:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "reference/evaluation.md",
        "line_start": 456,
        "line_end": 456,
        "snippet": "-u https://example.com/mcp \\"
      },
      {
        "id": "network:reference/evaluation.md:469:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "reference/evaluation.md",
        "line_start": 469,
        "line_end": 469,
        "snippet": "-u https://example.com/mcp \\"
      },
      {
        "id": "env_access:reference/evaluation.md:567:git-platform-tokens",
        "category": "env_access",
        "severity": "high",
        "pattern": "Git platform tokens",
        "file": "reference/evaluation.md",
        "line_start": 567,
        "line_end": 567,
        "snippet": "-e GITHUB_TOKEN=ghp_xxx \\"
      },
      {
        "id": "env_access:reference/evaluation.md:398:generic-api-secret-keys",
        "category": "env_access",
        "severity": "high",
        "pattern": "Generic API/secret keys",
        "file": "reference/evaluation.md",
        "line_start": 398,
        "line_end": 398,
        "snippet": "export ANTHROPIC_API_KEY=your_api_key_here"
      },
      {
        "id": "env_access:reference/evaluation.md:444:generic-api-secret-keys",
        "category": "env_access",
        "severity": "high",
        "pattern": "Generic API/secret keys",
        "file": "reference/evaluation.md",
        "line_start": 444,
        "line_end": 444,
        "snippet": "-e API_KEY=abc123 \\"
      },
      {
        "id": "env_access:reference/evaluation.md:557:generic-api-secret-keys",
        "category": "env_access",
        "severity": "high",
        "pattern": "Generic API/secret keys",
        "file": "reference/evaluation.md",
        "line_start": 557,
        "line_end": 557,
        "snippet": "export ANTHROPIC_API_KEY=your_api_key"
      },
      {
        "id": "blocker:reference/evaluation.md:364:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference/evaluation.md",
        "line_start": 364,
        "line_end": 364,
        "snippet": "Remember to parallelize solving tasks to avoid running out of context, then accumulate all answers a"
      },
      {
        "id": "blocker:reference/evaluation.md:547:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference/evaluation.md",
        "line_start": 547,
        "line_end": 547,
        "snippet": "<question>Find the project that was completed in December 2023 and had the longest duration from sta"
      },
      {
        "id": "blocker:reference/evaluation.md:374:network-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "Network reconnaissance",
        "file": "reference/evaluation.md",
        "line_start": 374,
        "line_end": 376,
        "snippet": "7. **Iterate and Refine** based on what you learn during the process"
      },
      {
        "id": "external_commands:reference/mcp_best_practices.md:223:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/mcp_best_practices.md",
        "line_start": 223,
        "line_end": 227,
        "snippet": "text: `Error: ${error.message}. Try using filter='active_only' to reduce results.`"
      },
      {
        "id": "blocker:reference/mcp_best_practices.md:27:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference/mcp_best_practices.md",
        "line_start": 27,
        "line_end": 27,
        "snippet": "- Avoid SSE (deprecated in favor of streamable HTTP)"
      },
      {
        "id": "blocker:reference/mcp_best_practices.md:54:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference/mcp_best_practices.md",
        "line_start": 54,
        "line_end": 54,
        "snippet": "4. **Be specific**: Avoid generic names that could conflict with other servers"
      },
      {
        "id": "blocker:reference/mcp_best_practices.md:235:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference/mcp_best_practices.md",
        "line_start": 235,
        "line_end": 235,
        "snippet": "- **Functional testing**: Verify correct execution with valid/invalid inputs"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:39:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 39,
        "line_end": 46,
        "snippet": "const output = { result: `Processed: ${param}` };"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:190:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 190,
        "line_end": 221,
        "snippet": "- Returns \"No users found matching '<query>'\" if search returns empty`,"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:221:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 221,
        "line_end": 247,
        "snippet": "text: `No users found matching '${params.query}'`"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:247:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 247,
        "line_end": 248,
        "snippet": "const lines = [`# User Search Results: '${params.query}'`, \"\","
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:248:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 248,
        "line_end": 250,
        "snippet": "`Found ${total} users (showing ${users.length})`, \"\"];"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:250:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 250,
        "line_end": 251,
        "snippet": "lines.push(`## ${user.name} (${user.id})`);"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:251:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 251,
        "line_end": 252,
        "snippet": "lines.push(`- **Email**: ${user.email}`);"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:252:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 252,
        "line_end": 274,
        "snippet": "if (user.team) lines.push(`- **Team**: ${user.team}`);"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:399:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 399,
        "line_end": 400,
        "snippet": "`Response truncated from ${data.length} to ${truncatedData.length} items. ` +"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:400:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 400,
        "line_end": 406,
        "snippet": "`Use 'offset' parameter or add filters to see more results.`;"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:426:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 426,
        "line_end": 432,
        "snippet": "return `Error: API request failed with status ${error.response.status}`;"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:432:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 432,
        "line_end": 434,
        "snippet": "return `Error: Unexpected error occurred: ${error instanceof Error ? error.message : String(error)}`"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:451:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 451,
        "line_end": 465,
        "snippet": "url: `${API_BASE_URL}/${endpoint}`,"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:474:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 474,
        "line_end": 480,
        "snippet": "const response = await axios.get(`${API_URL}/resource/${resourceId}`);"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:480:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 480,
        "line_end": 483,
        "snippet": "return axios.get(`${API_URL}/resource/${resourceId}`)"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:516:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 516,
        "line_end": 522,
        "snippet": "const data = await apiCall(`/users/${id}`);"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:522:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 522,
        "line_end": 524,
        "snippet": "return await apiCall(`/users/${id}`);  // No type safety"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:644:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 644,
        "line_end": 670,
        "snippet": "url: `${API_BASE_URL}/${endpoint}`,"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:670:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 670,
        "line_end": 676,
        "snippet": "return `Error: API request failed with status ${error.response.status}`;"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:676:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 676,
        "line_end": 690,
        "snippet": "return `Error: Unexpected error occurred: ${error instanceof Error ? error.message : String(error)}`"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:690:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 690,
        "line_end": 739,
        "snippet": "description: `[Full description as shown above]`,"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:739:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 739,
        "line_end": 756,
        "snippet": "console.error(`MCP server running on http://localhost:${port}/mcp`);"
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:802:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference/node_mcp_server.md",
        "line_start": 802,
        "line_end": 809,
        "snippet": "uri: `file://documents/${doc.name}`,"
      },
      {
        "id": "network:reference/node_mcp_server.md:413:http-client-library",
        "category": "network",
        "severity": "low",
        "pattern": "HTTP client library",
        "file": "reference/node_mcp_server.md",
        "line_start": 413,
        "line_end": 413,
        "snippet": "import axios, { AxiosError } from \"axios\";"
      },
      {
        "id": "network:reference/node_mcp_server.md:449:http-client-library",
        "category": "network",
        "severity": "low",
        "pattern": "HTTP client library",
        "file": "reference/node_mcp_server.md",
        "line_start": 449,
        "line_end": 449,
        "snippet": "const response = await axios({"
      },
      {
        "id": "network:reference/node_mcp_server.md:474:http-client-library",
        "category": "network",
        "severity": "low",
        "pattern": "HTTP client library",
        "file": "reference/node_mcp_server.md",
        "line_start": 474,
        "line_end": 474,
        "snippet": "const response = await axios.get(`${API_URL}/resource/${resourceId}`);"
      },
      {
        "id": "network:reference/node_mcp_server.md:480:http-client-library",
        "category": "network",
        "severity": "low",
        "pattern": "HTTP client library",
        "file": "reference/node_mcp_server.md",
        "line_start": 480,
        "line_end": 480,
        "snippet": "return axios.get(`${API_URL}/resource/${resourceId}`)"
      },
      {
        "id": "network:reference/node_mcp_server.md:548:http-client-library",
        "category": "network",
        "severity": "low",
        "pattern": "HTTP client library",
        "file": "reference/node_mcp_server.md",
        "line_start": 548,
        "line_end": 548,
        "snippet": "\"axios\": \"^1.7.9\","
      },
      {
        "id": "network:reference/node_mcp_server.md:598:http-client-library",
        "category": "network",
        "severity": "low",
        "pattern": "HTTP client library",
        "file": "reference/node_mcp_server.md",
        "line_start": 598,
        "line_end": 598,
        "snippet": "import axios, { AxiosError } from \"axios\";"
      },
      {
        "id": "network:reference/node_mcp_server.md:642:http-client-library",
        "category": "network",
        "severity": "low",
        "pattern": "HTTP client library",
        "file": "reference/node_mcp_server.md",
        "line_start": 642,
        "line_end": 642,
        "snippet": "const response = await axios({"
      },
      {
        "id": "network:reference/node_mcp_server.md:942:http-client-library",
        "category": "network",
        "severity": "low",
        "pattern": "HTTP client library",
        "file": "reference/node_mcp_server.md",
        "line_start": 942,
        "line_end": 942,
        "snippet": "- [ ] Error handling uses proper type guards (e.g., `axios.isAxiosError`, `z.ZodError`)"
      },
      {
        "id": "network:reference/node_mcp_server.md:424:python-http-libraries",
        "category": "network",
        "severity": "low",
        "pattern": "Python HTTP libraries",
        "file": "reference/node_mcp_server.md",
        "line_start": 424,
        "line_end": 424,
        "snippet": "return \"Error: Rate limit exceeded. Please wait before making more requests.\";"
      },
      {
        "id": "network:reference/node_mcp_server.md:668:python-http-libraries",
        "category": "network",
        "severity": "low",
        "pattern": "Python HTTP libraries",
        "file": "reference/node_mcp_server.md",
        "line_start": 668,
        "line_end": 668,
        "snippet": "return \"Error: Rate limit exceeded. Please wait before making more requests.\";"
      },
      {
        "id": "network:reference/node_mcp_server.md:601:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "reference/node_mcp_server.md",
        "line_start": 601,
        "line_end": 601,
        "snippet": "const API_BASE_URL = \"https://api.example.com/v1\";"
      },
      {
        "id": "network:reference/node_mcp_server.md:739:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "reference/node_mcp_server.md",
        "line_start": 739,
        "line_end": 739,
        "snippet": "console.error(`MCP server running on http://localhost:${port}/mcp`);"
      },
      {
        "id": "env_access:reference/node_mcp_server.md:707:environment-variable-access-dot-notation",
        "category": "env_access",
        "severity": "low",
        "pattern": "Environment variable access (dot notation)",
        "file": "reference/node_mcp_server.md",
        "line_start": 707,
        "line_end": 707,
        "snippet": "if (!process.env.EXAMPLE_API_KEY) {"
      },
      {
        "id": "env_access:reference/node_mcp_server.md:719:environment-variable-access-dot-notation",
        "category": "env_access",
        "severity": "low",
        "pattern": "Environment variable access (dot notation)",
        "file": "reference/node_mcp_server.md",
        "line_start": 719,
        "line_end": 719,
        "snippet": "if (!process.env.EXAMPLE_API_KEY) {"
      },
      {
        "id": "env_access:reference/node_mcp_server.md:737:environment-variable-access-dot-notation",
        "category": "env_access",
        "severity": "low",
        "pattern": "Environment variable access (dot notation)",
        "file": "reference/node_mcp_server.md",
        "line_start": 737,
        "line_end": 737,
        "snippet": "const port = parseInt(process.env.PORT || '3000');"
      },
      {
        "id": "env_access:reference/node_mcp_server.md:744:environment-variable-access-dot-notation",
        "category": "env_access",
        "severity": "low",
        "pattern": "Environment variable access (dot notation)",
        "file": "reference/node_mcp_server.md",
        "line_start": 744,
        "line_end": 744,
        "snippet": "const transport = process.env.TRANSPORT || 'stdio';"
      },
      {
        "id": "env_access:reference/node_mcp_server.md:707:environment-variable-object",
        "category": "env_access",
        "severity": "low",
        "pattern": "Environment variable object",
        "file": "reference/node_mcp_server.md",
        "line_start": 707,
        "line_end": 707,
        "snippet": "if (!process.env.EXAMPLE_API_KEY) {"
      },
      {
        "id": "env_access:reference/node_mcp_server.md:719:environment-variable-object",
        "category": "env_access",
        "severity": "low",
        "pattern": "Environment variable object",
        "file": "reference/node_mcp_server.md",
        "line_start": 719,
        "line_end": 719,
        "snippet": "if (!process.env.EXAMPLE_API_KEY) {"
      },
      {
        "id": "env_access:reference/node_mcp_server.md:737:environment-variable-object",
        "category": "env_access",
        "severity": "low",
        "pattern": "Environment variable object",
        "file": "reference/node_mcp_server.md",
        "line_start": 737,
        "line_end": 737,
        "snippet": "const port = parseInt(process.env.PORT || '3000');"
      },
      {
        "id": "env_access:reference/node_mcp_server.md:744:environment-variable-object",
        "category": "env_access",
        "severity": "low",
        "pattern": "Environment variable object",
        "file": "reference/node_mcp_server.md",
        "line_start": 744,
        "line_end": 744,
        "snippet": "const transport = process.env.TRANSPORT || 'stdio';"
      },
      {
        "id": "env_access:reference/node_mcp_server.md:707:generic-api-secret-keys",
        "category": "env_access",
        "severity": "high",
        "pattern": "Generic API/secret keys",
        "file": "reference/node_mcp_server.md",
        "line_start": 707,
        "line_end": 707,
        "snippet": "if (!process.env.EXAMPLE_API_KEY) {"
      },
      {
        "id": "env_access:reference/node_mcp_server.md:708:generic-api-secret-keys",
        "category": "env_access",
        "severity": "high",
        "pattern": "Generic API/secret keys",
        "file": "reference/node_mcp_server.md",
        "line_start": 708,
        "line_end": 708,
        "snippet": "console.error(\"ERROR: EXAMPLE_API_KEY environment variable is required\");"
      },
      {
        "id": "env_access:reference/node_mcp_server.md:719:generic-api-secret-keys",
        "category": "env_access",
        "severity": "high",
        "pattern": "Generic API/secret keys",
        "file": "reference/node_mcp_server.md",
        "line_start": 719,
        "line_end": 719,
        "snippet": "if (!process.env.EXAMPLE_API_KEY) {"
      },
      {
        "id": "env_access:reference/node_mcp_server.md:720:generic-api-secret-keys",
        "category": "env_access",
        "severity": "high",
        "pattern": "Generic API/secret keys",
        "file": "reference/node_mcp_server.md",
        "line_start": 720,
        "line_end": 720,
        "snippet": "console.error(\"ERROR: EXAMPLE_API_KEY environment variable is required\");"
      },
      {
        "id": "sensitive:reference/node_mcp_server.md:707:environment-file-access",
        "category": "sensitive",
        "severity": "high",
        "pattern": "Environment file access",
        "file": "reference/node_mcp_server.md",
        "line_start": 707,
        "line_end": 707,
        "snippet": "if (!process.env.EXAMPLE_API_KEY) {"
      },
      {
        "id": "sensitive:reference/node_mcp_server.md:719:environment-file-access",
        "category": "sensitive",
        "severity": "high",
        "pattern": "Environment file access",
        "file": "reference/node_mcp_server.md",
        "line_start": 719,
        "line_end": 719,
        "snippet": "if (!process.env.EXAMPLE_API_KEY) {"
      },
      {
        "id": "sensitive:reference/node_mcp_server.md:737:environment-file-access",
        "category": "sensitive",
        "severity": "high",
        "pattern": "Environment file access",
        "file": "reference/node_mcp_server.md",
        "line_start": 737,
        "line_end": 737,
        "snippet": "const port = parseInt(process.env.PORT || '3000');"
      },
      {
        "id": "sensitive:reference/node_mcp_server.md:744:environment-file-access",
        "category": "sensitive",
        "severity": "high",
        "pattern": "Environment file access",
        "file": "reference/node_mcp_server.md",
        "line_start": 744,
        "line_end": 744,
        "snippet": "const transport = process.env.TRANSPORT || 'stdio';"
      },
      {
        "id": "blocker:reference/node_mcp_server.md:102:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference/node_mcp_server.md",
        "line_start": 102,
        "line_end": 102,
        "snippet": "**Avoid Naming Conflicts**: Include the service context to prevent overlaps:"
      },
      {
        "id": "blocker:reference/node_mcp_server.md:289:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference/node_mcp_server.md",
        "line_start": 289,
        "line_end": 289,
        "snippet": ".email(\"Invalid email format\"),"
      },
      {
        "id": "blocker:reference/node_mcp_server.md:294:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference/node_mcp_server.md",
        "line_start": 294,
        "line_end": 294,
        "snippet": "}).strict();  // Use .strict() to forbid extra fields"
      },
      {
        "id": "blocker:reference/node_mcp_server.md:489:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference/node_mcp_server.md",
        "line_start": 489,
        "line_end": 489,
        "snippet": "3. **Avoid `any`**: Use proper types or `unknown` instead of `any`"
      },
      {
        "id": "blocker:reference/node_mcp_server.md:781:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference/node_mcp_server.md",
        "line_start": 781,
        "line_end": 781,
        "snippet": "throw new Error(\"Invalid URI format\");"
      },
      {
        "id": "blocker:reference/node_mcp_server.md:892:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference/node_mcp_server.md",
        "line_start": 892,
        "line_end": 892,
        "snippet": "2. **Avoid Duplication**:"
      },
      {
        "id": "network:reference/python_mcp_server.md:193:http-client-library",
        "category": "network",
        "severity": "low",
        "pattern": "HTTP client library",
        "file": "reference/python_mcp_server.md",
        "line_start": 193,
        "line_end": 193,
        "snippet": "data = await api_request(limit=params.limit, offset=params.offset)"
      },
      {
        "id": "network:reference/python_mcp_server.md:233:http-client-library",
        "category": "network",
        "severity": "low",
        "pattern": "HTTP client library",
        "file": "reference/python_mcp_server.md",
        "line_start": 233,
        "line_end": 233,
        "snippet": "async def _make_api_request(endpoint: str, method: str = \"GET\", **kwargs) -> dict:"
      },
      {
        "id": "network:reference/python_mcp_server.md:236:http-client-library",
        "category": "network",
        "severity": "low",
        "pattern": "HTTP client library",
        "file": "reference/python_mcp_server.md",
        "line_start": 236,
        "line_end": 236,
        "snippet": "response = await client.request("
      },
      {
        "id": "network:reference/python_mcp_server.md:382:http-client-library",
        "category": "network",
        "severity": "low",
        "pattern": "HTTP client library",
        "file": "reference/python_mcp_server.md",
        "line_start": 382,
        "line_end": 382,
        "snippet": "async def _make_api_request(endpoint: str, method: str = \"GET\", **kwargs) -> dict:"
      },
      {
        "id": "network:reference/python_mcp_server.md:385:http-client-library",
        "category": "network",
        "severity": "low",
        "pattern": "HTTP client library",
        "file": "reference/python_mcp_server.md",
        "line_start": 385,
        "line_end": 385,
        "snippet": "response = await client.request("
      },
      {
        "id": "network:reference/python_mcp_server.md:426:http-client-library",
        "category": "network",
        "severity": "low",
        "pattern": "HTTP client library",
        "file": "reference/python_mcp_server.md",
        "line_start": 426,
        "line_end": 426,
        "snippet": "data = await _make_api_request("
      },
      {
        "id": "network:reference/python_mcp_server.md:220:python-http-libraries",
        "category": "network",
        "severity": "low",
        "pattern": "Python HTTP libraries",
        "file": "reference/python_mcp_server.md",
        "line_start": 220,
        "line_end": 220,
        "snippet": "return \"Error: Rate limit exceeded. Please wait before making more requests.\""
      },
      {
        "id": "network:reference/python_mcp_server.md:260:python-http-libraries",
        "category": "network",
        "severity": "low",
        "pattern": "Python HTTP libraries",
        "file": "reference/python_mcp_server.md",
        "line_start": 260,
        "line_end": 260,
        "snippet": "response = requests.get(f\"{API_URL}/resource/{resource_id}\")  # Blocks"
      },
      {
        "id": "network:reference/python_mcp_server.md:402:python-http-libraries",
        "category": "network",
        "severity": "low",
        "pattern": "Python HTTP libraries",
        "file": "reference/python_mcp_server.md",
        "line_start": 402,
        "line_end": 402,
        "snippet": "return \"Error: Rate limit exceeded. Please wait before making more requests.\""
      },
      {
        "id": "network:reference/python_mcp_server.md:353:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "reference/python_mcp_server.md",
        "line_start": 353,
        "line_end": 353,
        "snippet": "API_BASE_URL = \"https://api.example.com/v1\""
      },
      {
        "id": "env_access:reference/python_mcp_server.md:511:generic-api-secret-keys",
        "category": "env_access",
        "severity": "high",
        "pattern": "Generic API/secret keys",
        "file": "reference/python_mcp_server.md",
        "line_start": 511,
        "line_end": 511,
        "snippet": "api_key = await ctx.elicit("
      },
      {
        "id": "env_access:reference/python_mcp_server.md:517:generic-api-secret-keys",
        "category": "env_access",
        "severity": "high",
        "pattern": "Generic API/secret keys",
        "file": "reference/python_mcp_server.md",
        "line_start": 517,
        "line_end": 517,
        "snippet": "return await api_call(resource_id, api_key)"
      },
      {
        "id": "blocker:reference/python_mcp_server.md:63:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference/python_mcp_server.md",
        "line_start": 63,
        "line_end": 63,
        "snippet": "**Avoid Naming Conflicts**: Include the service context to prevent overlaps:"
      },
      {
        "id": "blocker:reference/python_mcp_server.md:85:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference/python_mcp_server.md",
        "line_start": 85,
        "line_end": 85,
        "snippet": "extra='forbid'              # Forbid extra fields"
      },
      {
        "id": "blocker:reference/python_mcp_server.md:325:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference/python_mcp_server.md",
        "line_start": 325,
        "line_end": 325,
        "snippet": "- Returns \"Error: Invalid API authentication\" if API key is invalid (401 status)"
      },
      {
        "id": "blocker:reference/python_mcp_server.md:652:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference/python_mcp_server.md",
        "line_start": 652,
        "line_end": 652,
        "snippet": "2. **Avoid Duplication**:"
      },
      {
        "id": "blocker:reference/python_mcp_server.md:662:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference/python_mcp_server.md",
        "line_start": 662,
        "line_end": 662,
        "snippet": "3. **Avoid Manual Validation**: Let Pydantic handle input validation with constraints"
      },
      {
        "id": "blocker:reference/python_mcp_server.md:123:network-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "Network reconnaissance",
        "file": "reference/python_mcp_server.md",
        "line_start": 123,
        "line_end": 124,
        "snippet": "- Use `model_config` instead of nested `Config` class"
      },
      {
        "id": "sensitive:scripts/connections.py:80:environment-file-access",
        "category": "sensitive",
        "severity": "high",
        "pattern": "Environment file access",
        "file": "scripts/connections.py",
        "line_start": 80,
        "line_end": 80,
        "snippet": "self.env = env"
      },
      {
        "id": "sensitive:scripts/connections.py:84:environment-file-access",
        "category": "sensitive",
        "severity": "high",
        "pattern": "Environment file access",
        "file": "scripts/connections.py",
        "line_start": 84,
        "line_end": 84,
        "snippet": "StdioServerParameters(command=self.command, args=self.args, env=self.env)"
      },
      {
        "id": "external_commands:scripts/evaluation.py:204:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "scripts/evaluation.py",
        "line_start": 204,
        "line_end": 204,
        "snippet": "**Ground Truth Answer**: `{expected_answer}`"
      },
      {
        "id": "external_commands:scripts/evaluation.py:205:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "scripts/evaluation.py",
        "line_start": 205,
        "line_end": 205,
        "snippet": "**Actual Answer**: `{actual_answer}`"
      },
      {
        "id": "network:scripts/evaluation.py:315:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "scripts/evaluation.py",
        "line_start": 315,
        "line_end": 315,
        "snippet": "python evaluation.py -t sse -u https://example.com/mcp -H \"Authorization: Bearer token\" eval.xml"
      },
      {
        "id": "network:scripts/evaluation.py:318:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "scripts/evaluation.py",
        "line_start": 318,
        "line_end": 318,
        "snippet": "python evaluation.py -t http -u https://example.com/mcp -m claude-3-5-sonnet-20241022 eval.xml"
      },
      {
        "id": "sensitive:scripts/evaluation.py:344:environment-file-access",
        "category": "sensitive",
        "severity": "high",
        "pattern": "Environment file access",
        "file": "scripts/evaluation.py",
        "line_start": 344,
        "line_end": 344,
        "snippet": "env_vars = parse_env_vars(args.env) if args.env else None"
      },
      {
        "id": "blocker:scripts/evaluation.py:42:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "scripts/evaluation.py",
        "line_start": 42,
        "line_end": 42,
        "snippet": "- Comment on any errors encountered during tool usage: Did the tool fail to execute? Did the tool re"
      },
      {
        "id": "external_commands:SKILL.md:29:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 29,
        "line_end": 29,
        "snippet": "Clear, descriptive tool names help agents find the right tools quickly. Use consistent prefixes (e.g"
      },
      {
        "id": "external_commands:SKILL.md:41:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 41,
        "line_end": 41,
        "snippet": "Start with the sitemap to find relevant pages: `https://modelcontextprotocol.io/sitemap.xml`"
      },
      {
        "id": "external_commands:SKILL.md:43:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 43,
        "line_end": 43,
        "snippet": "Then fetch specific pages with `.md` suffix for markdown format (e.g., `https://modelcontextprotocol"
      },
      {
        "id": "external_commands:SKILL.md:61:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 61,
        "line_end": 61,
        "snippet": "- **TypeScript SDK**: Use WebFetch to load `https://raw.githubusercontent.com/modelcontextprotocol/t"
      },
      {
        "id": "external_commands:SKILL.md:65:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 65,
        "line_end": 65,
        "snippet": "- **Python SDK**: Use WebFetch to load `https://raw.githubusercontent.com/modelcontextprotocol/pytho"
      },
      {
        "id": "external_commands:SKILL.md:104:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 104,
        "line_end": 104,
        "snippet": "- Define `outputSchema` where possible for structured data"
      },
      {
        "id": "external_commands:SKILL.md:105:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 105,
        "line_end": 105,
        "snippet": "- Use `structuredContent` in tool responses (TypeScript SDK feature)"
      },
      {
        "id": "external_commands:SKILL.md:120:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 120,
        "line_end": 120,
        "snippet": "- `readOnlyHint`: true/false"
      },
      {
        "id": "external_commands:SKILL.md:121:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 121,
        "line_end": 121,
        "snippet": "- `destructiveHint`: true/false"
      },
      {
        "id": "external_commands:SKILL.md:122:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 122,
        "line_end": 122,
        "snippet": "- `idempotentHint`: true/false"
      },
      {
        "id": "external_commands:SKILL.md:123:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 123,
        "line_end": 123,
        "snippet": "- `openWorldHint`: true/false"
      },
      {
        "id": "external_commands:SKILL.md:140:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 140,
        "line_end": 140,
        "snippet": "- Run `npm run build` to verify compilation"
      },
      {
        "id": "external_commands:SKILL.md:141:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 141,
        "line_end": 141,
        "snippet": "- Test with MCP Inspector: `npx @modelcontextprotocol/inspector`"
      },
      {
        "id": "external_commands:SKILL.md:144:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 144,
        "line_end": 144,
        "snippet": "- Verify syntax: `python -m py_compile your_server.py`"
      },
      {
        "id": "external_commands:SKILL.md:184:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 184,
        "line_end": 192,
        "snippet": "```xml"
      },
      {
        "id": "external_commands:SKILL.md:192:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 192,
        "line_end": 203,
        "snippet": "```"
      },
      {
        "id": "external_commands:SKILL.md:203:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 203,
        "line_end": 203,
        "snippet": "- **MCP Protocol**: Start with sitemap at `https://modelcontextprotocol.io/sitemap.xml`, then fetch "
      },
      {
        "id": "external_commands:SKILL.md:212:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 212,
        "line_end": 213,
        "snippet": "- **Python SDK**: Fetch from `https://raw.githubusercontent.com/modelcontextprotocol/python-sdk/main"
      },
      {
        "id": "external_commands:SKILL.md:213:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 213,
        "line_end": 219,
        "snippet": "- **TypeScript SDK**: Fetch from `https://raw.githubusercontent.com/modelcontextprotocol/typescript-"
      },
      {
        "id": "external_commands:SKILL.md:219:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 219,
        "line_end": 226,
        "snippet": "- Tool registration with `@mcp.tool`"
      },
      {
        "id": "network:SKILL.md:41:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 41,
        "line_end": 41,
        "snippet": "Start with the sitemap to find relevant pages: `https://modelcontextprotocol.io/sitemap.xml`"
      },
      {
        "id": "network:SKILL.md:43:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 43,
        "line_end": 43,
        "snippet": "Then fetch specific pages with `.md` suffix for markdown format (e.g., `https://modelcontextprotocol"
      },
      {
        "id": "network:SKILL.md:61:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 61,
        "line_end": 61,
        "snippet": "- **TypeScript SDK**: Use WebFetch to load `https://raw.githubusercontent.com/modelcontextprotocol/t"
      },
      {
        "id": "network:SKILL.md:65:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 65,
        "line_end": 65,
        "snippet": "- **Python SDK**: Use WebFetch to load `https://raw.githubusercontent.com/modelcontextprotocol/pytho"
      },
      {
        "id": "network:SKILL.md:203:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 203,
        "line_end": 203,
        "snippet": "- **MCP Protocol**: Start with sitemap at `https://modelcontextprotocol.io/sitemap.xml`, then fetch "
      },
      {
        "id": "network:SKILL.md:212:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 212,
        "line_end": 212,
        "snippet": "- **Python SDK**: Fetch from `https://raw.githubusercontent.com/modelcontextprotocol/python-sdk/main"
      },
      {
        "id": "network:SKILL.md:213:hardcoded-url",
        "category": "network",
        "severity": "low",
        "pattern": "Hardcoded URL",
        "file": "SKILL.md",
        "line_start": 213,
        "line_end": 213,
        "snippet": "- **TypeScript SDK**: Fetch from `https://raw.githubusercontent.com/modelcontextprotocol/typescript-"
      }
    ],
    "finding_verdicts": [
      {
        "id": "network:reference/evaluation.md:456:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.94,
        "reason": "The URL is a placeholder or localhost example in documentation, not a hidden outbound request. It does not transmit user data by itself."
      },
      {
        "id": "network:reference/evaluation.md:469:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.94,
        "reason": "The URL is a placeholder or localhost example in documentation, not a hidden outbound request. It does not transmit user data by itself."
      },
      {
        "id": "env_access:reference/evaluation.md:567:git-platform-tokens",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched value is a placeholder credential in setup documentation, such as your_api_key or ghp_xxx. No real secret is embedded or read by the skill."
      },
      {
        "id": "env_access:reference/evaluation.md:398:generic-api-secret-keys",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched value is a placeholder credential in setup documentation, such as your_api_key or ghp_xxx. No real secret is embedded or read by the skill."
      },
      {
        "id": "env_access:reference/evaluation.md:444:generic-api-secret-keys",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched value is a placeholder credential in setup documentation, such as your_api_key or ghp_xxx. No real secret is embedded or read by the skill."
      },
      {
        "id": "env_access:reference/evaluation.md:557:generic-api-secret-keys",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched value is a placeholder credential in setup documentation, such as your_api_key or ghp_xxx. No real secret is embedded or read by the skill."
      },
      {
        "id": "blocker:reference/evaluation.md:364:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "blocker:reference/evaluation.md:547:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "blocker:reference/evaluation.md:374:network-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "external_commands:reference/mcp_best_practices.md:223:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "blocker:reference/mcp_best_practices.md:27:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "blocker:reference/mcp_best_practices.md:54:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "blocker:reference/mcp_best_practices.md:235:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:39:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:190:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:221:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:247:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:248:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:250:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:251:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:252:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:399:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:400:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:426:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:432:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:451:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:474:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:480:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:516:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:522:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:644:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:670:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:676:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:690:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:739:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:reference/node_mcp_server.md:802:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "network:reference/node_mcp_server.md:413:http-client-library",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/node_mcp_server.md:449:http-client-library",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/node_mcp_server.md:474:http-client-library",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/node_mcp_server.md:480:http-client-library",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/node_mcp_server.md:548:http-client-library",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/node_mcp_server.md:598:http-client-library",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/node_mcp_server.md:642:http-client-library",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/node_mcp_server.md:942:http-client-library",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/node_mcp_server.md:424:python-http-libraries",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/node_mcp_server.md:668:python-http-libraries",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/node_mcp_server.md:601:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.94,
        "reason": "The URL is a placeholder or localhost example in documentation, not a hidden outbound request. It does not transmit user data by itself."
      },
      {
        "id": "network:reference/node_mcp_server.md:739:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.94,
        "reason": "The URL is a placeholder or localhost example in documentation, not a hidden outbound request. It does not transmit user data by itself."
      },
      {
        "id": "env_access:reference/node_mcp_server.md:707:environment-variable-access-dot-notation",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The code reads example environment variables in a sample server to validate configuration. This is expected credential handling, not secret exfiltration."
      },
      {
        "id": "env_access:reference/node_mcp_server.md:719:environment-variable-access-dot-notation",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The code reads example environment variables in a sample server to validate configuration. This is expected credential handling, not secret exfiltration."
      },
      {
        "id": "env_access:reference/node_mcp_server.md:737:environment-variable-access-dot-notation",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The code reads example environment variables in a sample server to validate configuration. This is expected credential handling, not secret exfiltration."
      },
      {
        "id": "env_access:reference/node_mcp_server.md:744:environment-variable-access-dot-notation",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The code reads example environment variables in a sample server to validate configuration. This is expected credential handling, not secret exfiltration."
      },
      {
        "id": "env_access:reference/node_mcp_server.md:707:environment-variable-object",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The code reads example environment variables in a sample server to validate configuration. This is expected credential handling, not secret exfiltration."
      },
      {
        "id": "env_access:reference/node_mcp_server.md:719:environment-variable-object",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The code reads example environment variables in a sample server to validate configuration. This is expected credential handling, not secret exfiltration."
      },
      {
        "id": "env_access:reference/node_mcp_server.md:737:environment-variable-object",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The code reads example environment variables in a sample server to validate configuration. This is expected credential handling, not secret exfiltration."
      },
      {
        "id": "env_access:reference/node_mcp_server.md:744:environment-variable-object",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The code reads example environment variables in a sample server to validate configuration. This is expected credential handling, not secret exfiltration."
      },
      {
        "id": "env_access:reference/node_mcp_server.md:707:generic-api-secret-keys",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The code reads example environment variables in a sample server to validate configuration. This is expected credential handling, not secret exfiltration."
      },
      {
        "id": "env_access:reference/node_mcp_server.md:708:generic-api-secret-keys",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The code reads example environment variables in a sample server to validate configuration. This is expected credential handling, not secret exfiltration."
      },
      {
        "id": "env_access:reference/node_mcp_server.md:719:generic-api-secret-keys",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The code reads example environment variables in a sample server to validate configuration. This is expected credential handling, not secret exfiltration."
      },
      {
        "id": "env_access:reference/node_mcp_server.md:720:generic-api-secret-keys",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The code reads example environment variables in a sample server to validate configuration. This is expected credential handling, not secret exfiltration."
      },
      {
        "id": "sensitive:reference/node_mcp_server.md:707:environment-file-access",
        "verdict": "false_positive",
        "confidence": 0.87,
        "reason": "The match is process.env usage in reference code, not environment file access. It demonstrates expected configuration checks without exposing secrets."
      },
      {
        "id": "sensitive:reference/node_mcp_server.md:719:environment-file-access",
        "verdict": "false_positive",
        "confidence": 0.87,
        "reason": "The match is process.env usage in reference code, not environment file access. It demonstrates expected configuration checks without exposing secrets."
      },
      {
        "id": "sensitive:reference/node_mcp_server.md:737:environment-file-access",
        "verdict": "false_positive",
        "confidence": 0.87,
        "reason": "The match is process.env usage in reference code, not environment file access. It demonstrates expected configuration checks without exposing secrets."
      },
      {
        "id": "sensitive:reference/node_mcp_server.md:744:environment-file-access",
        "verdict": "false_positive",
        "confidence": 0.87,
        "reason": "The match is process.env usage in reference code, not environment file access. It demonstrates expected configuration checks without exposing secrets."
      },
      {
        "id": "blocker:reference/node_mcp_server.md:102:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "blocker:reference/node_mcp_server.md:289:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "blocker:reference/node_mcp_server.md:294:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "blocker:reference/node_mcp_server.md:489:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "blocker:reference/node_mcp_server.md:781:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "blocker:reference/node_mcp_server.md:892:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "network:reference/python_mcp_server.md:193:http-client-library",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/python_mcp_server.md:233:http-client-library",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/python_mcp_server.md:236:http-client-library",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/python_mcp_server.md:382:http-client-library",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/python_mcp_server.md:385:http-client-library",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/python_mcp_server.md:426:http-client-library",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/python_mcp_server.md:220:python-http-libraries",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/python_mcp_server.md:260:python-http-libraries",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/python_mcp_server.md:402:python-http-libraries",
        "verdict": "false_positive",
        "confidence": 0.9,
        "reason": "The match is tutorial code showing normal HTTP client usage for MCP integrations. It is transparent example code and not an unexpected network call by the skill."
      },
      {
        "id": "network:reference/python_mcp_server.md:353:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.94,
        "reason": "The URL is a placeholder or localhost example in documentation, not a hidden outbound request. It does not transmit user data by itself."
      },
      {
        "id": "env_access:reference/python_mcp_server.md:511:generic-api-secret-keys",
        "verdict": "false_positive",
        "confidence": 0.88,
        "reason": "The code demonstrates requesting a user-provided API key through an MCP context. It is documented sample behavior, not a hidden credential grab."
      },
      {
        "id": "env_access:reference/python_mcp_server.md:517:generic-api-secret-keys",
        "verdict": "false_positive",
        "confidence": 0.88,
        "reason": "The code demonstrates requesting a user-provided API key through an MCP context. It is documented sample behavior, not a hidden credential grab."
      },
      {
        "id": "blocker:reference/python_mcp_server.md:63:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "blocker:reference/python_mcp_server.md:85:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "blocker:reference/python_mcp_server.md:325:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "blocker:reference/python_mcp_server.md:652:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "blocker:reference/python_mcp_server.md:662:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "blocker:reference/python_mcp_server.md:123:network-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "sensitive:scripts/connections.py:80:environment-file-access",
        "verdict": "false_positive",
        "confidence": 0.82,
        "reason": "This stores and passes a caller-supplied environment dictionary to the MCP stdio subprocess; it does not read an environment file. The broader subprocess and env exposure risk is captured as a semantic finding."
      },
      {
        "id": "sensitive:scripts/connections.py:84:environment-file-access",
        "verdict": "false_positive",
        "confidence": 0.82,
        "reason": "This stores and passes a caller-supplied environment dictionary to the MCP stdio subprocess; it does not read an environment file. The broader subprocess and env exposure risk is captured as a semantic finding."
      },
      {
        "id": "external_commands:scripts/evaluation.py:204:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:scripts/evaluation.py:205:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "network:scripts/evaluation.py:315:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.94,
        "reason": "The URL is a placeholder or localhost example in documentation, not a hidden outbound request. It does not transmit user data by itself."
      },
      {
        "id": "network:scripts/evaluation.py:318:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.94,
        "reason": "The URL is a placeholder or localhost example in documentation, not a hidden outbound request. It does not transmit user data by itself."
      },
      {
        "id": "sensitive:scripts/evaluation.py:344:environment-file-access",
        "verdict": "false_positive",
        "confidence": 0.84,
        "reason": "This parses CLI-provided KEY=VALUE pairs and does not read files or discover secrets. The broader env-passthrough risk is captured as a semantic finding."
      },
      {
        "id": "blocker:scripts/evaluation.py:42:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The matched text is ordinary implementation or evaluation guidance, not reconnaissance or a blocker instruction. It does not gather system or network information."
      },
      {
        "id": "external_commands:SKILL.md:29:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:41:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:43:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:61:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:65:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:104:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:105:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:120:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:121:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:122:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:123:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:140:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:141:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:144:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:184:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:192:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:203:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:212:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:213:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "external_commands:SKILL.md:219:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This match is Markdown inline code or a TypeScript template literal in documentation, not shell execution. No command is invoked at this location."
      },
      {
        "id": "network:SKILL.md:41:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.93,
        "reason": "The URL points to MCP protocol or SDK documentation that the skill explicitly asks users to review. This is expected documentation access, not covert networking."
      },
      {
        "id": "network:SKILL.md:43:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.93,
        "reason": "The URL points to MCP protocol or SDK documentation that the skill explicitly asks users to review. This is expected documentation access, not covert networking."
      },
      {
        "id": "network:SKILL.md:61:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.93,
        "reason": "The URL points to MCP protocol or SDK documentation that the skill explicitly asks users to review. This is expected documentation access, not covert networking."
      },
      {
        "id": "network:SKILL.md:65:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.93,
        "reason": "The URL points to MCP protocol or SDK documentation that the skill explicitly asks users to review. This is expected documentation access, not covert networking."
      },
      {
        "id": "network:SKILL.md:203:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.93,
        "reason": "The URL points to MCP protocol or SDK documentation that the skill explicitly asks users to review. This is expected documentation access, not covert networking."
      },
      {
        "id": "network:SKILL.md:212:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.93,
        "reason": "The URL points to MCP protocol or SDK documentation that the skill explicitly asks users to review. This is expected documentation access, not covert networking."
      },
      {
        "id": "network:SKILL.md:213:hardcoded-url",
        "verdict": "false_positive",
        "confidence": 0.93,
        "reason": "The URL points to MCP protocol or SDK documentation that the skill explicitly asks users to review. This is expected documentation access, not covert networking."
      }
    ],
    "semantic_findings": [
      {
        "title": "Evaluation Harness Launches User-Supplied Commands",
        "description": "The evaluation script accepts a command and arguments, then the connection layer passes them to StdioServerParameters. Running untrusted evaluation settings could execute arbitrary local processes with supplied environment values.",
        "severity": "high",
        "locations": [
          {
            "file": "scripts/evaluation.py",
            "line_start": 327,
            "line_end": 329
          },
          {
            "file": "scripts/connections.py",
            "line_start": 76,
            "line_end": 84
          }
        ],
        "confidence": 0.92,
        "confidence_reasoning": "The CLI exposes command, args, and env options, and the connection code uses them to create a stdio MCP subprocess. This is direct local process launch behavior, although it is intended for trusted evaluations."
      },
      {
        "title": "Evaluation Data Sent To External LLM API",
        "description": "The evaluation harness sends questions, tool schemas, conversation history, and MCP tool results to Anthropic messages calls. Sensitive data returned by evaluated MCP tools could leave the local environment.",
        "severity": "medium",
        "locations": [
          {
            "file": "scripts/evaluation.py",
            "line_start": 17,
            "line_end": 17
          },
          {
            "file": "scripts/evaluation.py",
            "line_start": 96,
            "line_end": 103
          },
          {
            "file": "scripts/evaluation.py",
            "line_start": 116,
            "line_end": 120
          },
          {
            "file": "scripts/evaluation.py",
            "line_start": 137,
            "line_end": 144
          }
        ],
        "confidence": 0.88,
        "confidence_reasoning": "The script constructs Anthropic client calls with the evaluation messages and tools, and appends tool results into later requests. This clearly sends evaluation context to an external model API."
      },
      {
        "title": "Remote MCP Connections Accept Arbitrary URLs And Headers",
        "description": "The harness accepts user-provided SSE or HTTP URLs and headers, then connects using MCP remote transports. Tokens in headers could be sent to an unintended endpoint if copied from untrusted instructions.",
        "severity": "medium",
        "locations": [
          {
            "file": "scripts/evaluation.py",
            "line_start": 331,
            "line_end": 333
          },
          {
            "file": "scripts/connections.py",
            "line_start": 91,
            "line_end": 109
          }
        ],
        "confidence": 0.86,
        "confidence_reasoning": "The command-line interface accepts arbitrary remote URLs and headers, and the connection layer passes them directly to SSE or streamable HTTP clients. The risk depends on user-supplied configuration."
      }
    ],
    "critical_findings": [],
    "high_findings": [
      {
        "title": "Evaluation Harness Launches User-Supplied Commands",
        "description": "The evaluation script accepts a command and arguments, then the connection layer passes them to StdioServerParameters. Running untrusted evaluation settings could execute arbitrary local processes with supplied environment values.",
        "locations": [
          {
            "file": "scripts/evaluation.py",
            "line_start": 327,
            "line_end": 329
          },
          {
            "file": "scripts/connections.py",
            "line_start": 76,
            "line_end": 84
          }
        ],
        "review_kind": "security",
        "source_category": "semantic",
        "source_severity": "high",
        "confidence": 0.92,
        "confidence_reasoning": "The CLI exposes command, args, and env options, and the connection code uses them to create a stdio MCP subprocess. This is direct local process launch behavior, although it is intended for trusted evaluations."
      }
    ],
    "medium_findings": [
      {
        "title": "Evaluation Data Sent To External LLM API",
        "description": "The evaluation harness sends questions, tool schemas, conversation history, and MCP tool results to Anthropic messages calls. Sensitive data returned by evaluated MCP tools could leave the local environment.",
        "locations": [
          {
            "file": "scripts/evaluation.py",
            "line_start": 17,
            "line_end": 17
          },
          {
            "file": "scripts/evaluation.py",
            "line_start": 96,
            "line_end": 103
          },
          {
            "file": "scripts/evaluation.py",
            "line_start": 116,
            "line_end": 120
          },
          {
            "file": "scripts/evaluation.py",
            "line_start": 137,
            "line_end": 144
          }
        ],
        "review_kind": "security",
        "source_category": "semantic",
        "source_severity": "medium",
        "confidence": 0.88,
        "confidence_reasoning": "The script constructs Anthropic client calls with the evaluation messages and tools, and appends tool results into later requests. This clearly sends evaluation context to an external model API."
      },
      {
        "title": "Remote MCP Connections Accept Arbitrary URLs And Headers",
        "description": "The harness accepts user-provided SSE or HTTP URLs and headers, then connects using MCP remote transports. Tokens in headers could be sent to an unintended endpoint if copied from untrusted instructions.",
        "locations": [
          {
            "file": "scripts/evaluation.py",
            "line_start": 331,
            "line_end": 333
          },
          {
            "file": "scripts/connections.py",
            "line_start": 91,
            "line_end": 109
          }
        ],
        "review_kind": "security",
        "source_category": "semantic",
        "source_severity": "medium",
        "confidence": 0.86,
        "confidence_reasoning": "The command-line interface accepts arbitrary remote URLs and headers, and the connection layer passes them directly to SSE or streamable HTTP clients. The risk depends on user-supplied configuration."
      }
    ],
    "low_findings": [],
    "dangerous_patterns": [],
    "files_scanned": 10,
    "total_lines": 3532,
    "audit_model": "codex",
    "audited_at": "2026-07-05T15:29:47.815Z",
    "agent_auto_install_policy": "confirmation_required",
    "manual_install_policy": "allowed"
  },
  "content": {
    "user_title": "Build Reliable MCP Servers",
    "value_statement": "Building MCP servers can be slow because protocol details, schemas, transports, and evaluation design must align. This skill provides structured guidance, examples, and an evaluation harness for Python and TypeScript MCP integrations.",
    "seo_keywords": [
      "Claude",
      "Codex",
      "Claude Code",
      "MCP server",
      "Model Context Protocol",
      "FastMCP",
      "TypeScript MCP SDK",
      "AI tool integration",
      "MCP evaluation",
      "LLM tools"
    ],
    "actual_capabilities": [
      "Explains MCP server planning, transport selection, and tool design workflows.",
      "Provides TypeScript MCP SDK patterns with Zod validation and structured responses.",
      "Provides Python FastMCP patterns with Pydantic validation and async HTTP calls.",
      "Guides tool naming, pagination, error handling, annotations, and response formats.",
      "Includes an evaluation guide and scripts for testing MCP tools with Claude.",
      "Links to current MCP protocol and SDK documentation for deeper reference."
    ],
    "limitations": [
      "It is guidance material and examples, not a complete generated server.",
      "Examples use placeholder services, URLs, and credentials that must be replaced.",
      "The evaluation harness can run local commands and should only use trusted inputs.",
      "It does not manage hosting, deployment secrets, or production monitoring."
    ],
    "use_cases": [
      {
        "title": "Plan a New MCP Integration",
        "description": "Define tools, schemas, transports, and response formats before writing a server.",
        "target_user": "Application developer"
      },
      {
        "title": "Review Server Quality",
        "description": "Check an MCP implementation for naming, validation, pagination, error handling, and tool annotations.",
        "target_user": "Technical lead"
      },
      {
        "title": "Create Evaluation Tasks",
        "description": "Design read-only questions and run an evaluation harness to measure tool usefulness.",
        "target_user": "AI product engineer"
      }
    ],
    "prompt_templates": [
      {
        "title": "Plan a Starter Server",
        "scenario": "Beginner planning for a new integration.",
        "prompt": "Use mcp-builder to plan a basic MCP server for [service]. Include transport choice, tool names, schemas, response formats, and first test steps."
      },
      {
        "title": "Draft an Implementation",
        "scenario": "Intermediate implementation work in one language.",
        "prompt": "Use mcp-builder to draft a [TypeScript or Python] MCP server design for [service]. Focus on validation, pagination, error handling, and annotations."
      },
      {
        "title": "Design Evaluations",
        "scenario": "Advanced testing for a working server.",
        "prompt": "Use mcp-builder to create read-only evaluation tasks for my MCP server. Make each question stable, realistic, and easy to verify."
      },
      {
        "title": "Audit Production Readiness",
        "scenario": "Advanced review before release.",
        "prompt": "Use mcp-builder to review my MCP server plan for production readiness. Check security, authentication, schemas, tool descriptions, transport choice, and evaluation coverage."
      }
    ],
    "output_examples": [
      {
        "input": "Plan a GitHub issue tracking MCP server.",
        "output": "A phased implementation plan with tool names, required inputs, read-only annotations, pagination behavior, and evaluation ideas."
      },
      {
        "input": "Review a Python MCP server design for a CRM API.",
        "output": "A checklist covering Pydantic models, async HTTP usage, error messages, response formats, and missing security controls."
      },
      {
        "input": "Create evaluation questions for a support ticket MCP server.",
        "output": "Three stable read-only tasks with expected answer styles and notes about avoiding destructive operations."
      }
    ],
    "best_practices": [
      "Read current MCP protocol and SDK documentation before implementing server details.",
      "Use schema validation, pagination, clear annotations, and bounded responses for every tool.",
      "Run builds and evaluations only with trusted commands, credentials, URLs, and test data."
    ],
    "anti_patterns": [
      "Copying placeholder URLs, tokens, or service names into production code.",
      "Running evaluation commands from untrusted repositories or generated instructions.",
      "Returning unbounded API responses without filters, limits, or truncation messages."
    ],
    "faq": [
      {
        "question": "Does this skill create a complete MCP server automatically?",
        "answer": "No. It provides implementation guidance, examples, and review steps for building a server."
      },
      {
        "question": "Which languages does it cover?",
        "answer": "It covers TypeScript with the MCP SDK and Python with FastMCP."
      },
      {
        "question": "Does it require internet access?",
        "answer": "Some steps recommend fetching MCP documentation and SDK references from public documentation sites."
      },
      {
        "question": "Can it help with authentication design?",
        "answer": "Yes. It discusses API keys, OAuth guidance, environment variables, and clear authentication errors."
      },
      {
        "question": "Is the evaluation harness safe to run?",
        "answer": "Run it only with trusted commands, servers, URLs, headers, and credentials because it can launch processes and send data to Claude."
      },
      {
        "question": "Can I use it with Claude, Codex, and Claude Code?",
        "answer": "Yes. The skill declares support for Claude, Codex, and Claude Code workflows."
      }
    ]
  },
  "file_structure": [
    {
      "name": "LICENSE.txt",
      "type": "file",
      "path": "LICENSE.txt",
      "lines": 202
    },
    {
      "name": "reference",
      "type": "dir",
      "path": "reference",
      "children": [
        {
          "name": "evaluation.md",
          "type": "file",
          "path": "reference/evaluation.md",
          "lines": 602
        },
        {
          "name": "mcp_best_practices.md",
          "type": "file",
          "path": "reference/mcp_best_practices.md",
          "lines": 250
        },
        {
          "name": "node_mcp_server.md",
          "type": "file",
          "path": "reference/node_mcp_server.md",
          "lines": 970
        },
        {
          "name": "python_mcp_server.md",
          "type": "file",
          "path": "reference/python_mcp_server.md",
          "lines": 719
        }
      ]
    },
    {
      "name": "scripts",
      "type": "dir",
      "path": "scripts",
      "children": [
        {
          "name": "connections.py",
          "type": "file",
          "path": "scripts/connections.py",
          "lines": 152
        },
        {
          "name": "evaluation.py",
          "type": "file",
          "path": "scripts/evaluation.py",
          "lines": 374
        },
        {
          "name": "example_evaluation.xml",
          "type": "file",
          "path": "scripts/example_evaluation.xml",
          "lines": 23
        },
        {
          "name": "requirements.txt",
          "type": "file",
          "path": "scripts/requirements.txt",
          "lines": 3
        }
      ]
    },
    {
      "name": "SKILL.md",
      "type": "file",
      "path": "SKILL.md",
      "lines": 237
    }
  ]
}
brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).Stars: 322kUpdated Claude CodeCodexVerifiedInstall
brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).Stars: 198kUpdated Claude CodeCodexVerifiedInstall