LeanMCP Blog
Insights on development, design, and technology
Featured Post

We Got Tired of LLMs Getting LeanMCP Wrong. So We Built Our Own Agent
It took 2 hours. It runs on our own platform. And it knows our docs better than ChatGPT.
Explore by Topic
All Posts

WebMCP Solves Deployment. It Doesn't Solve Governance.
WebMCP shipped in Chrome 146 Canary two weeks ago. For deployment and authentication, the problem is largely solved. For governance and observability, the problem is just beginning.

MCP Apps: AI Agents Just Became Applications
MCP Apps (SEP-1865) is the open standard that lets AI agents render interactive UIs inside the conversation. Cross-platform, backed by Anthropic and OpenAI, live in Claude and ChatGPT today.

GPT Apps Are Here - Here's How to Build One in Production
OpenAI just opened the ChatGPT App Directory. If you're building with MCP, you're closer than you think. Here's everything you need to ship a GPT App that actually scales.

WashedMCP: One Query, Token-Efficient Semantic Code Search with Tool Call Automation
How WashedMCP won NexHacks by solving AI coding assistant inefficiencies: token-optimized semantic search, call-graph context, and automated tool memory with LeanMCP hooks.

LeanMCP: Run AI Agents in Production. Own Your Stack.
Open-Source Agent Runtime. Production-Grade Governance. Zero Vendor Lock-in. Getting an AI agent to work is easy. Getting it to production is a different story.

Announcing the LeanMCP Track Winners at NexHacks Carnegie Mellon!
Three outstanding projects demonstrated creative applications of LeanMCP - pushing the boundaries of context-efficient tool calling, semantic code understanding, and agentic AI systems.

TypeScript SDK for Building MCP Servers with type-safe decorators and streamable HTTP support
In 2025, MCP has matured as the open standard for connecting AI agents to external tools. LeanMCP SDK addresses production gaps with TypeScript decorators, compile-time safety, and zero-boilerplate deployment.
Launching Agent Runtime: Parallel MCP at Scale
AI agents need to make hundreds of MCP calls. Doing them sequentially is slow. Agent Runtime parallelizes everything. Here's how it works.
MCP vs Claude Skills: Infrastructure vs Instructions
MCP is production-grade infrastructure with real security and reliability. Claude Skills? Glorified RAG with Markdown files. Here's when you need each.
Releasing LeanMCP SDK
Introducing the LeanMCP SDK - built to solve the framework gap in MCP development with best practices, built-in tools, and seamless agent integration.