Conduit

What it does

Your agent got slower the more MCP servers you added, and it's not the model. Every server dumps its whole tool list into context on every request: 3 servers cost ~24k tokens before you even say hi. Conduit puts them behind one local gateway that exposes 3 meta-tools the agent searches on demand. Measured: 97% less tool overhead per request, ~90% fewer tokens, same task success. Cloud or local, one tool or five. Keys in your OS keychain. Free and open source.

A free, open source local MCP gateway. Set up each server once and every AI agent shares it (Claude, Cursor, VS Code, Codex), with up to 91% fewer tool tokens, secrets in your OS keychain, and one set of agent rules applied everywhere.

Set up your MCP servers once and share them across Claude, Cursor, Codex, and your other AI clients. Tools load on demand to keep context small. Your keys stay on your machine. Tool definitions, counted in tokens, load on every request before you've typed a word. Fewer tokens mean sharper answers, more headroom under your plan's limits, and lower bills. Those are tool-definition tokens on one request. Graded at the same task success, Toolport uses 74 to 91% fewer across servers · reproduce the benchmark Stop pasting the same server config and API key into Claude, then Cursor, then Codex. Set up and authenticate a server once in Toolport and it shows up in every agent, with hot toggles and…from toolport.app

• 
  Agent 37Mar 2026 · agent37.com · ▲401

  Hey HN, I noticed there are 1,000+ MCP skills on GitHub with zero way for creators to charge for them. So I built Agent37. Selling Claude skills today means asking customers to download your skill, set up Claude Code on their laptops, configure MCP servers, and pray it works. No trial, no updates, often you're just handing over the source. Agent37 lets creators upload a skill, share a link, and anyone can try it instantly (no Claude account needed). Built-in Stripe, creators keep 80%. Looking for feedback on the approach. Would you use something like this?

• 
  Mcp2cli – One CLI for every API, 96-99% fewer tokens than native MCPMar 2026 · github.com · ▲146

  Every MCP server injects its full tool schemas into context on every turn — 30 tools costs ~3,600 tokens/turn whether the model uses them or not. Over 25 turns with 120 tools, that's 362,000 tokens just for schemas. mcp2cli turns any MCP server or OpenAPI spec into a CLI at runtime. The LLM discovers tools on demand: mcp2cli --mcp https://mcp.example.com/sse --list # ~16 tokens/tool mcp2cli --mcp https://mcp.example.com/sse create-task --help # ~120 tokens, once mcp2cli --mcp https://mcp.example.com/sse create-task --title "Fix bug" No…

• 
  MCP3602025 · mcp360.ai · ▲61

  Hey HN Community, I built MCP360 after spending weeks integrating APIs for an AI agent project. Each tool needed its own setup, auth, billing, and maintenance. When any APIs changed, my integrations broke. I got tired of it. MCP360 is a single gateway giving AI agents access to 100+ tools through one config block. Search engines, web scraping, SEO, e-commerce data, maps, domain tools, and more. Real example: Instead of managing Google Search, web scraping, SERP tracking, and keyword research as 4 separate subscriptions (4 bills, 4 auth systems, 4 points of failure), you connect once. Works…

• 
  Remote OpenClawJul 2026 · codexmarketplaces.com · ▲81

  13,000+ MCP servers, skills & plugins for AI coding agents

• 
  Lazy-tool: reducing prompt bloat in MCP-based agent workflowsMar 2026 · github.com · ▲23

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

• 
  CrawlForge MCP11d ago · crawlforge.dev · ▲71

  Give your AI agent 31 web tools in one MCP server

• 
  Spira Maxima▲578

  Video model that turns script into viral social media videos

  AI · 18d ago · spira.ai

• 
  Rill Browser▲540

  The browser where Claude Code and Codex work beside you

  AI · 17d ago · rill.love

• 
  Germany's new sovereign AI model Kolibri▲424

  Kolibri is Aleph Alpha

  AI · 19d ago · tej.as

• 
  Giving Opus 5.5 a simulated paint canvas▲392

  Paintings by AI models: programs run through a simulation of oil paint.

  AI · 21d ago · stillwet.art

• 
  PixVerse R2▲381

  A real-time world model you can explore and change

  AI · 29d ago · world.pixverse.video

• 
  Hemory▲381

  Keep listening. Searchable memory for your AI agents.

  AI · 30d ago · hemory.com

• 
  Fundraisly▲1,537

  AI fundraising agent that finds investors and books meetings

  AI · Jun 2026 · fundraisly.com

• 
  Homebrew 6.0.0▲1,481

  Today, I’m proud to announce Homebrew 6.0.0. The most significant changes since 5.1.0 are a new tap trust security mechanism, the new faster, smaller, default internal Homebrew JSON API, sandboxing on Linux, better defaults informed by our user survey, many brew bundle improvements, improved performance and initial support for macOS 27 (Golden Gate). Happy to discuss any questions here!

  Dev tools · Jun 2026 · brew.sh

• 
  Performative-UI – A react component library of design tropes▲1,181

  Hi HN! I was super frustrated at apps like tiktok and wechat preventing me from saving or downloading the animated gifs, so I made a little tool to turn a screen recording into a slack-sized emoji. Hope you get some value from this!

  Life & fun · Jun 2026 · vorpus.github.io

• 
  Upstream▲976

  The inbox designed for humans and agents

  AI · Jun 2026 · upstream.do

• 
  Goldfish▲958

  Press Option. It knows your work and replies like you

  AI · Jun 2026 · goldfish.sh

• 
  I made Google Trends for Hacker News by indexing 18 years of comments▲822

  Search 18 years of Hacker News and chart how any topic, tool, or person trended. Overlay terms to compare their rise and fall across 45M posts and comments.

  Life & fun · Jun 2026 · hackernewstrends.com