MCPJam Inspector - Local Development and Testing Tool for MCP Servers
Research Date: January 26, 2026
Source URL: https://mcpjam.com
Documentation: https://docs.mcpjam.com
GitHub Repository: https://github.com/MCPJam/inspector
npm Package: https://www.npmjs.com/package/@mcpjam/inspector
Version at Research: v1.5.0
License: Apache-2.0
Overview
MCPJam Inspector is a comprehensive local development and testing tool for MCP servers, ChatGPT apps, and MCP Apps (ext-apps/SEP-1865). It provides a full widget emulator for ChatGPT and Claude app development, an LLM playground with free access to frontier models (GPT-5, Claude Sonnet), OAuth debugging, test case management with AI generation, and E2E evaluation capabilities via CLI.
Core Value Proposition: Eliminates the need for ngrok or ChatGPT subscriptions during MCP server and app development by providing a complete local testing environment with multi-LLM support.
Problem Addressed
| Problem |
How MCPJam Solves It |
| Complex MCP server setup for local testing |
Single npx command starts full inspector with web UI |
| ChatGPT app development requires ngrok + subscription |
Full widget emulator runs locally with no external dependencies |
| MCP Apps (Claude ext-apps) lack development tooling |
Complete SEP-1865 support with all JSON-RPC message types |
| OAuth 2.1 implementation debugging is complex |
Visual step-by-step OAuth debugger with multi-protocol support |
| Testing MCP tools across different LLMs is expensive |
Free access to GPT-5 and Claude Sonnet in playground |
| Manual test case creation is time-consuming |
AI-powered test generation from tool definitions |
| E2E testing MCP servers across clients is fragmented |
CLI-based eval framework simulates different client environments |
Key Statistics (as of January 26, 2026)
| Metric |
Value |
| GitHub Stars |
1,598 |
| Forks |
177 |
| Open Issues |
14 |
| Primary Language |
TypeScript |
| npm Package |
@mcpjam/inspector |
| CLI Package |
@mcpjam/cli |
| Latest Version |
v1.5.0 (Jan 20, 2026) |
| Created |
May 23, 2025 |
| Last Updated |
Jan 25, 2026 |
Key Features
1. MCP Server Inspection
- Multi-Server Connection: Connect to multiple MCP servers simultaneously
- Tool Testing: Manually invoke tools with custom parameters, view JSON-RPC logs
- Resource & Prompt Testing: Test resources, resource templates, and prompts
- Server Information: View icons, versions, capabilities, and instructions
- All JSON-RPC Logging: Real-time protocol message inspection
2. ChatGPT Apps SDK Support
- Full
window.openai API: Complete support for widgetState, callTool, structuredContent, sendFollowUpMessage, displayMode
- Widget Emulator: Preview widgets without ChatGPT subscription
- Device Emulation: Test across Desktop, Tablet, Mobile viewports
- CSP Testing: Validate Content Security Policy configurations
- Theme/Locale Testing: Light/dark modes, internationalization testing
- Safe Area Insets: Simulate device notches and gesture areas
3. MCP Apps (SEP-1865) Support
- Full JSON-RPC Message Types:
tools/call, ui/initialize, ui/message, ui/open-link
- Inline Widget Rendering: Render
ui.resourceUri components in chat
- Display Modes: Inline, Picture-in-Picture, Fullscreen support
4. OAuth Debugger
- Visual Step-by-Step: Interactive OAuth handshake walkthrough
- Multi-Protocol Versions: Supports specs 03-26, 06-18, and 11-25
- Registration Methods: Client pre-registration, DCR, and CIMD
- Network Inspection: View all HTTP requests/responses with headers
- Sequence Diagrams: Visual flow synchronized with progress
5. LLM Playground
- Free Frontier Models: GPT-5, Claude Sonnet, Gemini 2.5 at no cost
- Multi-Provider Support: OpenAI, Anthropic, Google, Deepseek, Mistral, Ollama, OpenRouter, LiteLLM
- Split-Panel Interface: Chat on left, JSON-RPC logs on right
- MCP Prompts: Use prompts directly via
/ command
- Elicitation Support: Handle interactive prompts from MCP servers
- Customizable Agent: System prompts, temperature settings
6. Test Cases
- AI-Powered Generation: Auto-generate tests from tool definitions using Claude Haiku
- Positive/Negative Tests: Test both tool triggering and non-triggering scenarios
- Multi-Model Evaluation: Run tests across different LLM providers
- Accuracy Metrics: Track pass rates, token consumption, duration
- Run History: Compare performance across runs and models
- App Store Compliance: Generate required test cases for ChatGPT app submission
7. MCP Evals (CLI)
- E2E Testing: Simulate end-user environments
- Environment Simulation: Test as Claude Desktop, Cursor, etc.
- Multiple Servers: Test against multiple MCP servers
- CI/CD Integration: JSON-based configuration for automation
Technical Architecture
MCPJam Inspector
|
+--------------------+--------------------+
| | |
Web UI CLI Tool Desktop App
(localhost:6274) (@mcpjam/cli) (Mac/Windows)
| | |
+--------------------+--------------------+
|
+-------------+-------------+
| | |
MCP Servers ChatGPT Apps MCP Apps
(stdio/HTTP) (window.openai) (SEP-1865)
| | |
+-------------+-------------+
|
+------+----+----+------+
| | | | |
OpenAI Claude Gemini Ollama ...
(GPT-5) (Sonnet) (Local)
Deployment Options
npx (Quick Start):
npx @mcpjam/inspector@latest
Docker:
docker run -p 127.0.0.1:6274:6274 mcpjam/mcp-inspector
Desktop App: Native Mac (.dmg) and Windows (.exe) installers available
With MCP Server:
npx @mcpjam/inspector@latest npx -y @modelcontextprotocol/server-everything
Configuration
Supports Claude Desktop-style configuration files:
{
"mcpServers": {
"weather-server": {
"command": "python",
"args": ["weather_server.py"],
"env": {
"WEATHER_API_KEY": "${WEATHER_API_KEY}"
}
}
}
}
Launch with: npx @mcpjam/inspector@latest --config path/to/config.json
Installation and Usage
Quick Start
# Start inspector
npx @mcpjam/inspector@latest
# Open browser to http://localhost:6274
Start with MCP Server
# Python FastMCP server
npx @mcpjam/inspector@latest uv run fastmcp run /path/to/server.py
# Node.js server
npx @mcpjam/inspector@latest npx -y your-mcp-package
Start with Ollama
npx @mcpjam/inspector@latest --ollama llama3.2
MCP Evals CLI
# Install CLI globally
npm install -g @mcpjam/cli
# Run evaluations
mcpjam evals run --tests tests.json --environment env.json
Test File Format
{
"tests": [
{
"title": "Test weather tool",
"query": "What's the weather in San Francisco?",
"expectedTools": ["get_weather"],
"model": { "id": "claude-3-5-sonnet-20241022", "provider": "anthropic" },
"selectedServers": ["weather-server"]
}
]
}
Comparison with Alternatives
| Feature |
MCPJam |
Official MCP Inspector |
Postman |
Custom Testing |
| MCP Server Testing |
Yes |
Yes |
Partial |
Manual |
| ChatGPT Apps Support |
Yes |
No |
No |
Manual |
| MCP Apps (SEP-1865) |
Yes |
No |
No |
Manual |
| OAuth Debugger |
Yes (visual) |
Basic |
Yes |
Manual |
| LLM Playground |
Yes (free models) |
No |
No |
Requires API keys |
| AI Test Generation |
Yes |
No |
No |
No |
| E2E Evals CLI |
Yes |
No |
No |
Manual |
| Widget Emulator |
Full |
No |
No |
Manual |
| Multi-Provider LLMs |
8+ providers |
N/A |
N/A |
Per-provider setup |
Relevance to Claude Code Development
Direct Applications
- MCP Server Development: Test custom MCP servers before integrating with Claude Code
- OAuth Flow Debugging: Debug authentication for MCP servers requiring OAuth
- Tool Testing: Manually invoke and verify tool behavior before deployment
- Multi-LLM Validation: Verify MCP servers work across Claude, GPT, Gemini models
Patterns Worth Adopting
- Visual OAuth Debugging: Step-by-step handshake visualization pattern
- AI Test Generation: Using LLMs to generate test cases from tool definitions
- E2E Eval Framework: JSON-based test configuration for CI/CD integration
- Free Frontier Model Access: Proxied access pattern for development tools
- Widget Emulation: Iframe isolation with injected APIs for UI testing
- Multi-Protocol Support: Backwards-compatible OAuth spec handling
Integration Opportunities
- Claude Code Plugin: Create plugin that launches MCPJam for connected MCP servers
- Pre-Deployment Testing: Integrate MCPJam evals into plugin CI/CD pipelines
- OAuth Debugging Workflow: Use MCPJam for debugging MCP OAuth implementations
- Test Case Library: Generate and maintain test cases for Claude Code MCP integrations
Complementary to Existing MCP Ecosystem Research
MCPJam fills a testing and development tooling gap:
- Narsil MCP: Code intelligence and security scanning (runtime)
- OctoCode MCP: Research Driven Development with GitHub search (research)
- Docs MCP Server: Documentation grounding and retrieval (knowledge)
- MCPJam: Development, testing, and evaluation (DevOps/QA)
References
- Documentation: https://docs.mcpjam.com (accessed 2026-01-26)
- GitHub Repository: https://github.com/MCPJam/inspector (accessed 2026-01-26)
- npm Package (@mcpjam/inspector): https://www.npmjs.com/package/@mcpjam/inspector (accessed 2026-01-26)
- npm Package (@mcpjam/cli): https://www.npmjs.com/package/@mcpjam/cli (accessed 2026-01-26)
- Getting Started Guide: https://docs.mcpjam.com/getting-started.md (accessed 2026-01-26)
- OAuth Debugger: https://docs.mcpjam.com/inspector/guided-oauth.md (accessed 2026-01-26)
- App Builder: https://docs.mcpjam.com/inspector/app-builder.md (accessed 2026-01-26)
- LLM Playground: https://docs.mcpjam.com/inspector/llm-playground.md (accessed 2026-01-26)
- Test Cases: https://docs.mcpjam.com/inspector/test-cases.md (accessed 2026-01-26)
- MCP Evals: https://docs.mcpjam.com/evals/overview.md (accessed 2026-01-26)
- MCP Protocol Specification: https://modelcontextprotocol.io/ (accessed 2026-01-26)
- ChatGPT Apps SDK: https://developers.openai.com/apps-sdk/ (accessed 2026-01-26)
- MCP Apps SEP-1865: https://github.com/modelcontextprotocol/modelcontextprotocol/pull/1865 (accessed 2026-01-26)
Freshness Tracking
| Field |
Value |
| Last Verified |
2026-01-26 |
| Version at Verification |
v1.5.0 |
| GitHub Stars at Verification |
1,598 |
| CLI Version at Verification |
v1.1.6 |
| Next Review Recommended |
2026-04-26 (3 months) |
Change Detection Indicators:
- Monitor GitHub releases for version changes
- Check npm for new versions of @mcpjam/inspector and @mcpjam/cli
- Track SEP-1865 (MCP Apps) specification evolution
- Review changelog for new LLM provider support
- Monitor for new OAuth protocol version support
- Track ChatGPT Apps SDK changes that may affect widget emulation
1---2name: mcpjam-inspector-local-development-and-testing-tool-for-mcp3description: MCPJam Inspector is a comprehensive local development and testing tool for MCP servers, ChatGPT apps, and MCP Apps (ext-apps/SEP-1865).4---5# MCPJam Inspector - Local Development and Testing Tool for MCP Servers67**Research Date**: January 26, 20268**Source URL**: <https://mcpjam.com>9**Documentation**: <https://docs.mcpjam.com>10**GitHub Repository**: <https://github.com/MCPJam/inspector>11**npm Package**: <https://www.npmjs.com/package/@mcpjam/inspector>12**Version at Research**: v1.5.013**License**: Apache-2.01415---1617## Overview1819MCPJam Inspector is a comprehensive local development and testing tool for MCP servers, ChatGPT apps, and MCP Apps (ext-apps/SEP-1865). It provides a full widget emulator for ChatGPT and Claude app development, an LLM playground with free access to frontier models (GPT-5, Claude Sonnet), OAuth debugging, test case management with AI generation, and E2E evaluation capabilities via CLI.2021**Core Value Proposition**: Eliminates the need for ngrok or ChatGPT subscriptions during MCP server and app development by providing a complete local testing environment with multi-LLM support.2223---2425## Problem Addressed2627| Problem | How MCPJam Solves It |28| ----------------------------------------------------- | ---------------------------------------------------------------- |29| Complex MCP server setup for local testing | Single npx command starts full inspector with web UI |30| ChatGPT app development requires ngrok + subscription | Full widget emulator runs locally with no external dependencies |31| MCP Apps (Claude ext-apps) lack development tooling | Complete SEP-1865 support with all JSON-RPC message types |32| OAuth 2.1 implementation debugging is complex | Visual step-by-step OAuth debugger with multi-protocol support |33| Testing MCP tools across different LLMs is expensive | Free access to GPT-5 and Claude Sonnet in playground |34| Manual test case creation is time-consuming | AI-powered test generation from tool definitions |35| E2E testing MCP servers across clients is fragmented | CLI-based eval framework simulates different client environments |3637---3839## Key Statistics (as of January 26, 2026)4041| Metric | Value |42| ---------------- | --------------------- |43| GitHub Stars | 1,598 |44| Forks | 177 |45| Open Issues | 14 |46| Primary Language | TypeScript |47| npm Package | @mcpjam/inspector |48| CLI Package | @mcpjam/cli |49| Latest Version | v1.5.0 (Jan 20, 2026) |50| Created | May 23, 2025 |51| Last Updated | Jan 25, 2026 |5253---5455## Key Features5657### 1. MCP Server Inspection5859- **Multi-Server Connection**: Connect to multiple MCP servers simultaneously60- **Tool Testing**: Manually invoke tools with custom parameters, view JSON-RPC logs61- **Resource & Prompt Testing**: Test resources, resource templates, and prompts62- **Server Information**: View icons, versions, capabilities, and instructions63- **All JSON-RPC Logging**: Real-time protocol message inspection6465### 2. ChatGPT Apps SDK Support6667- **Full `window.openai` API**: Complete support for `widgetState`, `callTool`, `structuredContent`, `sendFollowUpMessage`, `displayMode`68- **Widget Emulator**: Preview widgets without ChatGPT subscription69- **Device Emulation**: Test across Desktop, Tablet, Mobile viewports70- **CSP Testing**: Validate Content Security Policy configurations71- **Theme/Locale Testing**: Light/dark modes, internationalization testing72- **Safe Area Insets**: Simulate device notches and gesture areas7374### 3. MCP Apps (SEP-1865) Support7576- **Full JSON-RPC Message Types**: `tools/call`, `ui/initialize`, `ui/message`, `ui/open-link`77- **Inline Widget Rendering**: Render `ui.resourceUri` components in chat78- **Display Modes**: Inline, Picture-in-Picture, Fullscreen support7980### 4. OAuth Debugger8182- **Visual Step-by-Step**: Interactive OAuth handshake walkthrough83- **Multi-Protocol Versions**: Supports specs 03-26, 06-18, and 11-2584- **Registration Methods**: Client pre-registration, DCR, and CIMD85- **Network Inspection**: View all HTTP requests/responses with headers86- **Sequence Diagrams**: Visual flow synchronized with progress8788### 5. LLM Playground8990- **Free Frontier Models**: GPT-5, Claude Sonnet, Gemini 2.5 at no cost91- **Multi-Provider Support**: OpenAI, Anthropic, Google, Deepseek, Mistral, Ollama, OpenRouter, LiteLLM92- **Split-Panel Interface**: Chat on left, JSON-RPC logs on right93- **MCP Prompts**: Use prompts directly via `/` command94- **Elicitation Support**: Handle interactive prompts from MCP servers95- **Customizable Agent**: System prompts, temperature settings9697### 6. Test Cases9899- **AI-Powered Generation**: Auto-generate tests from tool definitions using Claude Haiku100- **Positive/Negative Tests**: Test both tool triggering and non-triggering scenarios101- **Multi-Model Evaluation**: Run tests across different LLM providers102- **Accuracy Metrics**: Track pass rates, token consumption, duration103- **Run History**: Compare performance across runs and models104- **App Store Compliance**: Generate required test cases for ChatGPT app submission105106### 7. MCP Evals (CLI)107108- **E2E Testing**: Simulate end-user environments109- **Environment Simulation**: Test as Claude Desktop, Cursor, etc.110- **Multiple Servers**: Test against multiple MCP servers111- **CI/CD Integration**: JSON-based configuration for automation112113---114115## Technical Architecture116117```text118 MCPJam Inspector119 |120 +--------------------+--------------------+121 | | |122 Web UI CLI Tool Desktop App123(localhost:6274) (@mcpjam/cli) (Mac/Windows)124 | | |125 +--------------------+--------------------+126 |127 +-------------+-------------+128 | | |129 MCP Servers ChatGPT Apps MCP Apps130 (stdio/HTTP) (window.openai) (SEP-1865)131 | | |132 +-------------+-------------+133 |134 +------+----+----+------+135 | | | | |136 OpenAI Claude Gemini Ollama ...137 (GPT-5) (Sonnet) (Local)138```139140### Deployment Options141142**npx (Quick Start)**:143144```bash145npx @mcpjam/inspector@latest146```147148**Docker**:149150```bash151docker run -p 127.0.0.1:6274:6274 mcpjam/mcp-inspector152```153154**Desktop App**: Native Mac (.dmg) and Windows (.exe) installers available155156**With MCP Server**:157158```bash159npx @mcpjam/inspector@latest npx -y @modelcontextprotocol/server-everything160```161162### Configuration163164Supports Claude Desktop-style configuration files:165166```json167{168 "mcpServers": {169 "weather-server": {170 "command": "python",171 "args": ["weather_server.py"],172 "env": {173 "WEATHER_API_KEY": "${WEATHER_API_KEY}"174 }175 }176 }177}178```179180Launch with: `npx @mcpjam/inspector@latest --config path/to/config.json`181182---183184## Installation and Usage185186### Quick Start187188```bash189# Start inspector190npx @mcpjam/inspector@latest191192# Open browser to http://localhost:6274193```194195### Start with MCP Server196197```bash198# Python FastMCP server199npx @mcpjam/inspector@latest uv run fastmcp run /path/to/server.py200201# Node.js server202npx @mcpjam/inspector@latest npx -y your-mcp-package203```204205### Start with Ollama206207```bash208npx @mcpjam/inspector@latest --ollama llama3.2209```210211### MCP Evals CLI212213```bash214# Install CLI globally215npm install -g @mcpjam/cli216217# Run evaluations218mcpjam evals run --tests tests.json --environment env.json219```220221### Test File Format222223```json224{225 "tests": [226 {227 "title": "Test weather tool",228 "query": "What's the weather in San Francisco?",229 "expectedTools": ["get_weather"],230 "model": { "id": "claude-3-5-sonnet-20241022", "provider": "anthropic" },231 "selectedServers": ["weather-server"]232 }233 ]234}235```236237---238239## Comparison with Alternatives240241| Feature | MCPJam | Official MCP Inspector | Postman | Custom Testing |242| -------------------- | ----------------- | ---------------------- | ------- | ------------------ |243| MCP Server Testing | Yes | Yes | Partial | Manual |244| ChatGPT Apps Support | Yes | No | No | Manual |245| MCP Apps (SEP-1865) | Yes | No | No | Manual |246| OAuth Debugger | Yes (visual) | Basic | Yes | Manual |247| LLM Playground | Yes (free models) | No | No | Requires API keys |248| AI Test Generation | Yes | No | No | No |249| E2E Evals CLI | Yes | No | No | Manual |250| Widget Emulator | Full | No | No | Manual |251| Multi-Provider LLMs | 8+ providers | N/A | N/A | Per-provider setup |252253---254255## Relevance to Claude Code Development256257### Direct Applications2582591. **MCP Server Development**: Test custom MCP servers before integrating with Claude Code2602. **OAuth Flow Debugging**: Debug authentication for MCP servers requiring OAuth2613. **Tool Testing**: Manually invoke and verify tool behavior before deployment2624. **Multi-LLM Validation**: Verify MCP servers work across Claude, GPT, Gemini models263264### Patterns Worth Adopting2652661. **Visual OAuth Debugging**: Step-by-step handshake visualization pattern2672. **AI Test Generation**: Using LLMs to generate test cases from tool definitions2683. **E2E Eval Framework**: JSON-based test configuration for CI/CD integration2694. **Free Frontier Model Access**: Proxied access pattern for development tools2705. **Widget Emulation**: Iframe isolation with injected APIs for UI testing2716. **Multi-Protocol Support**: Backwards-compatible OAuth spec handling272273### Integration Opportunities2742751. **Claude Code Plugin**: Create plugin that launches MCPJam for connected MCP servers2762. **Pre-Deployment Testing**: Integrate MCPJam evals into plugin CI/CD pipelines2773. **OAuth Debugging Workflow**: Use MCPJam for debugging MCP OAuth implementations2784. **Test Case Library**: Generate and maintain test cases for Claude Code MCP integrations279280### Complementary to Existing MCP Ecosystem Research281282MCPJam fills a testing and development tooling gap:283284- **Narsil MCP**: Code intelligence and security scanning (runtime)285- **OctoCode MCP**: Research Driven Development with GitHub search (research)286- **Docs MCP Server**: Documentation grounding and retrieval (knowledge)287- **MCPJam**: Development, testing, and evaluation (DevOps/QA)288289---290291## References2922931. **Documentation**: <https://docs.mcpjam.com> (accessed 2026-01-26)2942. **GitHub Repository**: <https://github.com/MCPJam/inspector> (accessed 2026-01-26)2953. **npm Package (@mcpjam/inspector)**: <https://www.npmjs.com/package/@mcpjam/inspector> (accessed 2026-01-26)2964. **npm Package (@mcpjam/cli)**: <https://www.npmjs.com/package/@mcpjam/cli> (accessed 2026-01-26)2975. **Getting Started Guide**: <https://docs.mcpjam.com/getting-started.md> (accessed 2026-01-26)2986. **OAuth Debugger**: <https://docs.mcpjam.com/inspector/guided-oauth.md> (accessed 2026-01-26)2997. **App Builder**: <https://docs.mcpjam.com/inspector/app-builder.md> (accessed 2026-01-26)3008. **LLM Playground**: <https://docs.mcpjam.com/inspector/llm-playground.md> (accessed 2026-01-26)3019. **Test Cases**: <https://docs.mcpjam.com/inspector/test-cases.md> (accessed 2026-01-26)30210. **MCP Evals**: <https://docs.mcpjam.com/evals/overview.md> (accessed 2026-01-26)30311. **MCP Protocol Specification**: <https://modelcontextprotocol.io/> (accessed 2026-01-26)30412. **ChatGPT Apps SDK**: <https://developers.openai.com/apps-sdk/> (accessed 2026-01-26)30513. **MCP Apps SEP-1865**: <https://github.com/modelcontextprotocol/modelcontextprotocol/pull/1865> (accessed 2026-01-26)306307---308309## Freshness Tracking310311| Field | Value |312| ---------------------------- | --------------------- |313| Last Verified | 2026-01-26 |314| Version at Verification | v1.5.0 |315| GitHub Stars at Verification | 1,598 |316| CLI Version at Verification | v1.1.6 |317| Next Review Recommended | 2026-04-26 (3 months) |318319**Change Detection Indicators**:320321- Monitor GitHub releases for version changes322- Check npm for new versions of @mcpjam/inspector and @mcpjam/cli323- Track SEP-1865 (MCP Apps) specification evolution324- Review changelog for new LLM provider support325- Monitor for new OAuth protocol version support326- Track ChatGPT Apps SDK changes that may affect widget emulation