nowfound

AI · May 7, 2025

CF

Cloi – free local debugging agent in your terminal

Hey everyone! For the past two weeks my friend and I have been heads-down building Cloi, a fully local debugging agent that runs right in your terminal. You probably know the drill—every AI coding tool asks for API keys, subscriptions, and uploads your entire codebase to the cloud. Cloi does none of that: it runs entirely on your machine, with no cloud, no API keys, no subscriptions, and zero data leaving your system. What Cloi does: - Contextual error capture: Grabs your stack trace, local files, and environment to understand the issue. - Local LLM inference: Spins up Ollama on your box and…

What it does

In the maker’s words, at launch

Hey everyone! For the past two weeks my friend and I have been heads-down building Cloi, a fully local debugging agent that runs right in your terminal. You probably know the drill—every AI coding tool asks for API keys, subscriptions, and uploads your entire codebase to the cloud. Cloi does none of that: it runs entirely on your machine, with no cloud, no API keys, no subscriptions, and zero data leaving your system. What Cloi does: - Contextual error capture: Grabs your stack trace, local files, and environment to understand the issue. - Local LLM inference: Spins up Ollama on your box and generates targeted fixes—no external servers. - Safe patch application: Presents you with diffs and only applies changes when you explicitly approve. - Model‐agnostic: Ships with Phi-4 out of the box (surprisingly capable for its size!), but you can swap in any Ollama model you’ve installed. Why we built it: - Maintain full control over your code and data—ideal for security-sensitive projects - Avoid recurring subscription fees and cloud vendor lock-in - Keep your development flow entirely offline when you need it Highlights: We hit 202 stars in just 5 days, which tells us we're not the only ones who wanted this! Cloi is plug-and-play (just install and run), and we designed it to be completely unopinionated, meaning you can you whatever Ollama model you want. Get it now: npm install -g @cloi-ai/cloi If you find Cloi useful, we’d really appreciate a star on GitHub. Try it out, let us know what you think, and happy debugging! — Gabriel Cha & Mingyou Kim

Does the same job

all alternatives →
  • agents-cliJul 2026 · github.com · ▲145

    The CLI your coding agent uses to ship agents

  • ReplicasJun 2026 · ▲239

    Run Claude Code and Codex in the cloud

  • Raindrop WorkshopMay 2026 · ▲189

    Open source, free, local debugger for AI agents

  • Epho16d ago · epho.io · ▲156

    Run Claude Code, Codex or Opencode in cloud with your repo

  • kimiflareApr 2026 · ▲96

    kimi k2.6 cli code editor hosted on cloudflare workers AI

  • Bob's CLIJun 2026 · ▲151

    A local-first AI coding CLI that adapts to you

More ai this month

the category →
  • I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.

    AI · 17d ago · simedw.com

  • Astute585

    Automate your B2B brand going viral, with new media creators

    AI · 18d ago · company-app.joinastute.com

  • Grok Bot547

    AI teammates that you can give real work to

    AI · 25d ago · x.ai

  • Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…

    AI · 27d ago · cactuscompute.com

  • Make your software self-driving

    AI · 30d ago · coldtea.ai

  • Soloop472

    Approval-first Agent OS for solo founders

    AI · 30d ago · soloop.io

Launched alongside, May 2025

the whole month →