MCP Studio v2.2.0
Get Early Bird Access
v2.2.0 Release • Desktop IDE for Model Context Protocol

Desktop IDE and AI Simulator for Model Context Protocol

Inspect, simulate, benchmark, test, and package MCP servers. Connect local stdio processes, remote SSE endpoints, and LLM agents in a visual workspace.

Download Free Community Edition
No monthly fees (Bring Your Own Key and offline Ollama support)
Windows, macOS & Linux
Compatible with official MCP SDK specifications
MCP Studio v2.2.0: Protocol IDE
Ollama local: connected (llama3:8b) 15 Active QA Tools
query_database SQL Tool
Tool Arguments (Schema Form)
sql: "SELECT id, email, status FROM users WHERE active = true LIMIT 10;"
timeout_ms: 5000
Response (Smart Table) 200 OK • 14ms
✓ 3 rows returned
[ { id: 1, user: "alexander", role: "admin" } ]
AI Agent Simulator: Ollama (llama3:8b) automatically invoked query_database and synthesized natural language answer.
242 Tokens • $0.00

Compatible with standard Model Context Protocol clients and frameworks

Claude Desktop Cursor IDE Windsurf Local Ollama OpenAI GPT-4o Anthropic Claude 3.7 Python FastMCP TypeScript SDK

Built for AI Engineers and Tool Developers

Inspect, Test, and Ship MCP Servers

Inspect tool definitions visually, generate mock payloads, run automated assertions, and test agent behavior with local or cloud models instead of parsing raw JSON-RPC logs.

Dynamic JSON Schema Form Runner

Translates tool parameter schemas into interactive forms with type validation, environment variable interpolation ({{API_KEY}}), and one-click execution.

Smart Mock Data Auto-Filler

Generates realistic test values for parameter names like emails, SQL queries, UUIDs, and file paths in one click without manual typing.

Multi-View Response Visualizer

Inspect responses through interactive collapsible JSON trees, searchable tables for database rows, rendered Markdown, and decoded media previews.

Local Ollama & Cloud LLM Simulator

Test how models invoke your tools using local Ollama (llama3:8b) or cloud providers (Claude 3.7, GPT-4o, Gemini 2.0). Review tool selection and reasoning steps directly.

AI Model Arena Shootout

Run comparative benchmarks between two different models on the same tools to compare tool selection, latency, and response accuracy side by side.

Live JSON-RPC 2.0 Packet Stream

Inspect raw request and response frames with latency timers, set breakpoint interceptors to modify parameters during execution, and replay past calls.

Developer Toolkit

15 Built-in Tools for Protocol Development

Testing, code scaffolding, and containerization utilities in a single desktop application.

01
Assertion Test Suites
Automated regression testing with status, latency, and payload assertions.
02
Concurrency & Latency Benchmark
Measure tool throughput, error rates, and p50, p95, and p99 latency percentiles under load.
03
Security & Sandbox Auditor
Static vulnerability scanner detecting API key leaks, path traversal risks, and injection vulnerabilities in tool schemas.
04
In-Memory Mock MCP Server
Run local mock servers with custom JSON fixtures and configurable latency delays.
05
Visual Workflow Pipelines
Chain multiple tool calls together and pass dynamic arguments between steps (e.g. {{steps[0].result.id}}).
06
Server Project Scaffolder
Generate starter projects with boilerplate code for Python (FastMCP), TypeScript, or Go MCP servers in one click.
07
OpenAPI to MCP Transpiler
Convert OpenAPI specifications directly into executable MCP tool definitions.
08
Schema Diff & Changelog
Track schema changes across versions and export clean Markdown changelogs.
09
Interactive API Documentation
Generate interactive documentation in Markdown and standalone HTML formats in one click.
10
Process Watchdog Supervisor
Real-time RSS memory, CPU, and uptime monitoring with auto-recovery on crashes.
11
MCP Gateway & Breakpoints
Set JSON-RPC breakpoints to pause, inspect, and modify tool parameters before execution.
12
Docker & Cloud Packager
Generate multi-stage Dockerfiles, Docker Compose files, and Kubernetes manifests.
13
Remote SSE Tunnel & Bridge
Connect remote SSE endpoints and expose local stdio servers over secure public tunnels.
14
Session Recorder & Replay
Export and import portable .mcpsession bundles to share and reproduce bugs.
15
Multi-Language Code Export
Generate ready-to-run client snippets in Python, TypeScript, cURL, Go, and Rust.
Product Launch Deal • 50% Off

Simple, Transparent Pricing

No recurring subscriptions. Lifetime licenses include one full year of feature updates.

Community Edition

Free Forever

Essential inspection and testing for individual developers.

$0 Free
1 Active Server Connection
Dynamic Form Runner & Schema Validation
Smart Mock Data Auto-Filler
Resource Viewer & Prompt Tester
Live JSON-RPC Traffic Stream
Multi-Language Code Snippets (5 Languages)
Download Free Edition
Early Bird • 50% Off

Lifetime Pro License

Includes the complete developer toolkit, multi-model simulation arena, and unlimited servers.

$49.00 $24.99 One-Time License
Unlimited Active Server Workspace Tabs
Multi-LLM AI Agent Simulator (Ollama + Cloud)
AI Model Arena Shootout (Side-by-Side Dual LLM)
Full 15-Tool Developer Toolkit Access
Automated Assertion Test Suites & Regressions
Visual Workflow Multi-Tool Chaining
In-Memory Mock MCP Server & Fixtures Engine
OpenAPI Transpiler & Docker/K8s Packager
1-Click Cursor, Claude & Windsurf Exporters
Offline License Verification + 1 Year of Feature Updates

Processed securely via Paddle • Instant license key delivery

Instant Delivery: Your license key is cryptographically tied to your specified email and will be emailed immediately upon checkout. You can activate it directly inside MCP Studio desktop.

Frequently Asked Questions

Do I need to pay monthly subscription fees for LLMs?

No. MCP Studio operates on a Bring Your Own Key (BYOK) model. For cloud models (Claude, OpenAI, Gemini), you provide your own API key and pay the provider directly based on usage. If you run local Ollama (llama3:8b), AI simulation runs completely offline at zero cost.

How does auto-discovery work?

MCP Studio automatically reads existing configuration files from Claude Desktop (claude_desktop_config.json) and Cursor IDE (mcp.json) so you can inspect active servers without re-entering environment variables.

Can I test remote HTTP / SSE MCP servers?

Yes. MCP Studio supports both local sub-processes (stdio via Node, Python, UV, Docker, or binaries) and remote Server-Sent Events (SSE endpoints).

What is included with the Pro Lifetime License?

The Pro license unlocks unlimited server tabs, the multi-LLM simulation engine, the AI Model Arena, and all 15 developer tools, plus 1 year of all feature updates and offline license validation.