More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    MCP Doctor — Protocol Health & Schema Linter

    1

    Audit, lint and verify MCP server configurations, tool collisions, missing binaries, and JSON-RPC handshakes.

    Free

    1 useSecurity scanned
    MCP Doctor — Protocol Health & Schema Linter

    MCP Doctor — Protocol Health & Schema Linter

    Example session with this skill installed

    Run MCP Doctor on my Claude Desktop configuration and verify live handshake connectivity.

    • Read your context and instructions
    • Compiled the mcp doctor

    ================================================================
    🩺 MCP DOCTOR — Health & Schema Linter (v1.0.0)

    Overall Status: 🟢 HEALTHY
    Configs Scanned: 1 | Servers Audited: 2
    Summary: 🔴 0 Critical | 🟡 0 Warnings | 🟢 2 Passed

    🟢 [OK] Server: [filesystem]
    Details: Handshake probe passed: Handshake successful (JSON-RPC responded)

    🟢 [OK] Server: [git-tools]
    Details: Handshake probe passed: Handshake successful (JSON-RPC responded)

    All MCP servers validated. Zero tool collisions detected.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    MCP Doctor — Model Context Protocol Health & Schema Linter

    Frustrated by AI coding assistants silently failing when invoking Model Context Protocol (MCP) servers due to subtle JSON syntax errors, missing executables on PATH, or tool name collisions?

    MCP Doctor is a production-grade diagnostic CLI and schema linter for the open Model Context Protocol standard. It inspects your Claude Desktop, Cursor, Windsurf, Antigravity, and custom agent configurations in milliseconds.

    ⚡ Key Capabilities:

    • Schema & Syntax Linting: Validates JSON compliance and ensures required command, args, and env attributes match the official MCP specification.
    • Binary & PATH Resolution: Instantly detects if uvx, npx, python3, or custom binaries are missing or inaccessible before an agent crashes.
    • Credential & Placeholder Watchdog: Flags unconfigured API keys and dummy placeholder tokens (YOUR_KEY_HERE, sk-ant-xxx).
    • Cross-Config Collision Detection: Identifies duplicated server names across multiple configuration files.
    • Live JSON-RPC Handshake Probe: Performs a non-destructive stdio handshake (initialize call with strict timeout) to prove connectivity without side effects.
    • Agent-Ready JSON Output: Seamlessly feeds structured diagnostics to autonomous agent orchestrators.

    📦 What's Included:

    • scripts/mcp_doctor.py — Production CLI diagnostic linter
    • scripts/test_mcp_doctor.py — Automated test suite with 100% pass rate
    • references/mcp_doctor_guide.md — CLI reference & troubleshooting matrix
    • references/json_rpc_schema_spec.md — Formal 2024-11-05 MCP spec summary
    • SKILL.md — Lean Core prompt rules (< 40 lines) for Claude Code, Cursor, and Antigravity

    🔒 100% Local-First & Zero Telemetry

    Runs entirely on your local machine. No external dependencies, zero cloud egress.

    Requires Python 3. Cannot auto-repair broken binaries or permission errors; diagnostic only. Handshake tests require the server process to be executable.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 29 days ago

    • Free with an Agensi account

    Requires

    python3
    Terminal
    Read Files

    Permissions

    Terminal / Shell
    Read Files
    Listed29 days ago

    What's inside

    Frequently Asked Questions

    MCP Doctor acts as a diagnostic tool that scans your system to identify why tools might be failing or missing. It validates JSON-RPC connectivity, checks for missing binaries in your PATH, and flags tool name collisions that prevent agents from functioning correctly.

    The skill is designed to audit configurations for Claude Desktop, Cursor, and Antigravity. It is compatible with any environment that follows the official Model Context Protocol (MCP) JSON-RPC specification.

    You will receive the core diagnostic tool, a comprehensive setup guide, and example configurations to help you interpret scan results. It also includes the CLI-based script ready for immediate system-wide auditing.

    Yes, while it automatically scans standard paths for popular agents, you can use the `--config` flag to point the tool at any specific JSON configuration file, making it ideal for custom development workflows.

    You need Python 3 installed on your host machine to run the diagnostics. While the tool identifies permission issues and missing dependencies, it requires you to have administrative access to manually fix system-level path or permission errors it discovers.

    New in LLM Ops