Back to browse
GitHub Repository

local-first MCP discovery runtime for agents — search before invoke, reduce prompt bloat, and route to local MCP tools

26 starsGo

Lazy-tool: reducing prompt bloat in MCP-based agent workflows

by like-to-code1·Mar 31, 2026·23 points·1 comment

AI Analysis

●●●BangerBig BrainZero to One

Cuts MCP prompt tokens 46% with single Go binary, no Docker or vector DB.

Strengths
  • Auto-imports existing Claude Desktop, Cursor, and VS Code MCP configurations
  • Three modes—direct, search, hybrid—switchable via command-line flag
  • Circuit breaking, caching, and session reuse built into proxy layer
Weaknesses
  • Claude Code's built-in tool search covers single-provider workflows already
  • Only 5 stars suggests very early adoption and untested at scale
Category
Target Audience

Developers using MCP servers with AI coding assistants

Similar To

RAG-MCP · MetaMCP · AWS AgentCore Gateway

Similar Projects

AI/ML●●Solid

Zenflow a multi-agent orchestration and workflow engine

Race-safe mailboxes for agent coordination is a clever specific touch.

Big BrainShip It
vietanh85
201mo ago