wasmcp for MCP
Wasmcp: A WebAssembly Kit for Polyglot Model Context Protocol Servers
- Free
- 4.9
- V cli-v0.4.13
Wasmcp: A WebAssembly Kit for Polyglot Model Context Protocol Servers
Wasmcp is a WebAssembly component development kit for the Model Context Protocol (MCP), built with Rust. It enables developers to create polyglot MCP servers that connect to AI clients and Large Language Models. These servers act as a foundation for tool calling and context provision within distributed intelligent systems. Wasmcp focuses on building scalable, agent-facing applications, allowing integration of diverse programming languages within a single WebAssembly component. It targets WebAssembly as its runtime, ensuring portability across various deployment environments.
What does Wasmcp connect to?
Wasmcp functions as a WebAssembly component development kit for the Model Context Protocol (MCP). It establishes connections with AI clients and Large Language Models that support the MCP standard. The tool acts as a server, providing capabilities for tool calling and context provision to these connected AI entities. It is designed to integrate with various agent SDKs and inference APIs, including the OpenAI Agents SDK, Google's Agent Development Kit, and the OpenAI Responses API. Compatible AI clients include ChatGPT/Codex, Claude Code/Desktop, and Gemini CLI.
Core features and tools
MCP servers built with Wasmcp expose several server features, including tools, resources, prompts, and completions. These servers can also call client-hosted features, such as elicitations. Specific callable tools within individual components handle requests like "tools/call". Wasmcp supports stdio for local subprocess communication and Streamable HTTP for multiple clients, sessions, and authorization. Custom transport protocols can also be implemented to extend its connectivity. Authentication models include public access by default or OAuth 2.1 protection.
Who is it for?
Wasmcp targets application developers and DevOps engineers who aim to construct scalable, agent-facing applications. It is particularly valuable for those building distributed intelligent systems where the Model Context Protocol forms the underlying communication standard. This includes individuals needing to mix tools and features written in various languages like Rust, Python, or TypeScript into a single WebAssembly component. Skip this if your project does not involve WebAssembly components or the Model Context Protocol.
How to get started
Access is granted through a standard OAuth authorization flow or is public by default. For OAuth 2.1, set WASMCP_AUTH_MODE to "oauth" and configure JWT_ISSUER and JWT_JWKS_URI environment variables.
Use cases
- Building composable MCP servers by mixing tools and features written in various languages (e.g., Rust, Python, TypeScript) into a single WebAssembly component binary.
- Deploying sandboxed MCP server binaries locally, on network edges (e.g., Fermyon Wasm Functions), or on Kubernetes clusters (e.g., SpinKube).
- Implementing polyglot MCP servers with diverse toolsets, such as Python tools using Pandas, TypeScript tools using Zod, and Rust for performance-critical or authorization middleware.
Wasmcp provides a specialized kit for developing Model Context Protocol servers using WebAssembly. It supports polyglot component creation and deployment across diverse environments, from local systems to Kubernetes. The platform offers flexibility in authentication and integrates with established AI clients and agent SDKs. Its focus on agent-facing applications makes it suitable for distributed intelligent systems.
Pros
- Enables polyglot MCP server development.
- Supports multiple AI clients and SDKs.
- Offers flexible authentication options.
- Deploys across various environments.
Cons
- SSE transport support is not confirmed.
- OAuth setup requires multiple environment variables.
- Building from source needs Cargo installation.
- Install script fetches binary from external URL.
wasmcp for MCP
- Free
- 4.9
- V cli-v0.4.13
Top downloads

aiope
Aiope: An MCP Server for AI-Powered Development and Integrations
unbrowser
unbrowser: an agent‑native, Chrome‑free runtime for LLM workflows

Task-Scheduler-Studio
WinUI 3 task manager with MCP AI integration for Windows

DataNexa
DataNexa: A Secure AI Database Gateway
Little-Claude
Little-Claude desktop GUI for Claude Code with MCP integration
Discover more programs
onemcp-hub
- 4.1
- Free
Central Hub for OneMCP User Contributions

touchdesigner-mcp
- 4.6
- Free
Touchdesigner MCP: AI Bridge for Creative TouchDesigner Control
flow-like
- 4.2
- Free
Flow Like: A Visual Platform for AI, Workflows, and Data Integration

dify-mcp-client
- 4
- Free
Dify MCP Client: Connecting Dify to External Model Context Protocol Servers
pctx
- 4.2
- Free
Pctx: An Execution Layer for AI Agentic Tool Calls
paiml-mcp-agent-toolkit
- 4.2
- Free
Paiml MCP Agent Toolkit: AI-Powered Code Analysis for Developers

tuui
- 4.6
- Free
TUUI: A Comprehensive Local AI Playground

mcp-servers-for-revit
- 4.9
- Free
Integrate AI with Autodesk Revit Effortlessly

narsil-mcp
- 4
- Free
Comprehensive AI Code Intelligence Tool

browser-control-mcp
- 4.8
- Free
Browser Control MCP: AI-Driven Firefox Automation for Agents