Best for
- Individuals wanting a locally running AI in the browser
- Developers interested in voice interaction and MCP integration
J
JARVIS MCP Server
Community
Not a standard server
Your private, personal AI assistant with voice interaction and real-time chat.
This is not a typical connect-and-use MCP server; read the source repository before using it.
- Category
- Other #179 of 230
- Stars
- ★ 81 Popular
- Transport
- stdio (local process)
- Runtime
- Node.js 18+
- Credentials
- No credential needed
- License
- MIT
- Last commit
45FMRS · D
JARVIS offers an integrated voice and chat experience, but as a personal project, official support is limited, making it suitable for experimentation and personal use.
Strongest · Setup experience 11/20 Weakest · Reliability 8/20
Why each score
- Reliability 8/20
- The codebase includes full TypeScript source, package.json scripts, and Vite config. The MCP server module exports a server instance and registers tool lists. However, no tests or CI configuration are present, so actual runtime correctness cannot be verified, leading to a moderate score.
- Security and permissions 8/20
- No obvious malicious behavior or credential leaks were found in the source. API keys are managed via environment variables for external services, but there is no evidence of least-privilege implementation, confirmation for dangerous operations, or explicit data-flow disclosure.
- Maintenance 9/20
- The repository has recent commits (last commit around Feb 2025) and a MIT license, but no releases, dependency update history, or dedicated security response channel are evident.
- Documentation 9/20
- README provides clear installation, usage, and MCP configuration instructions, including scripts and feature overview. However, it lacks detailed tool parameter documentation, troubleshooting guides, and cost/limit information.
- Setup experience 11/20
- Setup is straightforward with Node.js 18+ and a modern browser. But no client configuration examples are given, and reliance on WebGPU may limit some environments.
Static review · not runListed 2026-08-07
Read the FMRS scoring method →Fit and risk
Not for
- Users needing cloud-deployed production environments
- Users without WebGPU support or modern browsers
Required permissions
- Requires browser WebGPU support for local model execution
- May require microphone access for voice input
Risks and side effects
- Local browser execution may consume significant computational resources
- Data privacy risks when relying on external MCP servers
Setup
Before you start
Runtime：Node.js 18+
- Clone the repository: git clone <repository-url>
- Navigate into the directory and install dependencies: npm install
- Start the development server: npm run dev
- Open your browser to http://localhost:8080.
Check that it works
Run npm run dev and open http://localhost:8080; if the app loads and the /mcp page lets you add MCP servers, the install works.
Troubleshooting
- Ensure Node.js version >= 18; browser must support WebGPU; check microphone permissions.
Things to try
Once connected, you can ask your AI assistant things like:
- Open http://localhost:8080 and switch to JARVIS voice mode to talk with the assistant
Use cases
Voice-controlled personal assistant
Real-time text chat
Image content description extraction
Overview
JARVIS is a modern AI assistant web application featuring voice interaction, real-time conversation, and MCP (Model Context Protocol) server integration. It runs AI models locally in the browser using Transformers.js, supporting voice activity detection, speech-to-text, text-to-speech, and image-to-text capabilities. Users can interact via voice or traditional chat, and configure MCP servers for extended functionality.
Similar servers
Hevy MCP
77 · B
Manage your Hevy workout data from AI assistants via MCP.
Firecrawl MCP Server
75 · B
Firecrawl's official MCP server for web search, scraping, and structured extraction for AI agents
BioMCP
75 · B
One binary. One grammar. Evidence from the biomedical sources you already trust.
Compartment
75 · B
Encrypted, fully offline long-term memory for AI agents
Source revision 3b553fcc8f7d Data synced 2026-10-11 Read the FMRS scoring method