Best Open Source Tools (2026)

A newspaper-style Hacker News reader that shows articles and comments side by side. Dark mode, installable PWA, shareable story links, and a free weekly newsletter. Open-source.

AI agents run on your machine, with your credentials, making network requests you never see. Greywall puts a deny-by-default layer in front of any CLI agent. Run greywall -- claude and it restricts file access, blocks outbound connections, and shows you a live dashboard of what the agent is trying to reach. You decide what to allow. Nothing is containerized. Your workflow stays the same but safer. It runs on Linux and macOS and costs nothing.

Visualize your LLM workflows as interactive & shareable graphs. Point Codag at your codebase — it maps every LLM call, branch, and data flow. Click any node to jump to source. Live updates as you code. Self-hosted, open source.

The Desktop Buddy sits persistently in the corner of your screen as a small animated gecko. Click it once, describe what you need, and it handles it - without you switching windows, opening new tabs, or navigating menus. Skales is a local first AI desktop agent for Windows and macOS. It manages emails, calendar, browser tasks and research autonomously. Asks before anything destructive. Works with any LLM: OpenRouter, Anthropic, Gemini or fully offline via Ollama. No Docker - No terminal.

Browzer is an AI-powered workflow automation tool that learns your repetitive tasks and processes to automate them with 1 prompt. Record once & automate forever. Browzer doesn't assume your processes like other agentic tools & browsers, but instead understands your processes to automate work, the way you'd do it.

Holdor is a free, open-source macOS menu bar app that prevents sleep while Claude, Cursor, Windsurf, VS Code, and other AI coding agents are running. Lock your screen, walk away, agents keep working.

ScreenForge — Free, Open-Source Screen Time for Windows It helps you understand where your time actually goes. It’s a native Windows app that tracks app usage, reveals habits, and turns activity into clear daily and weekly insights. Everything runs locally, no accounts, no cloud, no tracking. Lightweight, quiet, and customizable with multiple themes. Built with Electron and React. Free forever and open source under MIT License.

Clawdboard is a free, open-source leaderboard for Claude Code usage. Run one command (npx clawdboard) to sync your usage, then see how your spend, tokens, streaks, and model mix stack up against other developers. No prompts or code ever leave your machine. What you get: personal profiles with usage charts and badges, team competition, and shareable stats. 100% built with Claude Code by a growth marketer. Full codebase on GitHub for self-hosting, forking, or contribution

Bank statement parsers are hard to test. Real data is private, every bank formats differently, and everyone claims "99% accuracy" with no way to verify. The Bankstatemently Open Benchmark is a free dataset of synthetic bank and credit card statement PDFs designed around real parsing challenges. 5 countries, 14 edge cases, fictional banks with realistic formatting. Download, parse, submit to the free evaluation API, and get field-level accuracy scores. MIT-licensed, open-source on GitHub.

Agent Settlement Extension (ASE) is an economic metadata layer that extends existing A2A and MCP agent communication protocols with financial semantics. It adds standardized structures for cost attribution, delegation, and settlement events, enabling agents to exchange not just tasks, but economic intent. ASE is part of the Caracal project, building the foundation for governed, auditable agent economies.

A lightweight alternative to OpenClaw that runs in containers for security. OpenClaw has 400,000 lines of code. NanoClaw does the same thing in 4000. NanoClaw is a personal Claude assistant that runs on WhatsApp and Telegram. It manages your email, researches the web, runs scheduled briefings, and coordinates agent swarms, all from your phone. The difference? Every agent runs inside an isolated Linux container. No ambient access. Secure by default

Bootstrap team. No VC. We fine-tuned our own AI model and open sourced it. LumiChat is built around it with an agentic mode that writes and executes real Node.js code in a sandboxed browser (no server), persistent memory, Study Mode, and RAG for large documents. Multi-model support. ₹69/day. The model is ours, the code is open, and the agent is extensible by anyone.

WebVM is a virtual Linux environment running in the browser via WebAssembly. WebVM doesn’t require registration. It can be forked, unmodified with no restrictions. WebVM is powered by CheerpX, an x86 virtualization engine in WASM which enables the safe, sandboxed execution of Linux binaries client-side. CheerpX includes an x86-to-WebAssembly JIT compiler, a virtual block-based file system, and a Linux syscall emulator. Both WebVM and CheerpX are free-to-use for FOSS and personal projects.

grep was built in 1973. It finds text, not meaning. When you search for "authentication logic" but the code is named handleUserSession, grep fails. grepai uses vector embeddings to understand code semantically. Search by WHAT code does, not what it's called. Built for AI agents like Claude Code and Cursor, find relevant code instantly instead of browsing hundreds of files. Features: • Semantic search with natural language • Call graph tracing • 100% local • Real-time file watching

Shear adds real cut-and-paste file actions to macOS Finder. Use ⌘X to cut files and ⌘V to move them instantly. Just one of the missing features macOS users have wanted for years!

YourClaw is built for the OpenClaw ecosystem. We’ve eliminated the "Docker drama" so you can deploy, skill, and scale a collaborative team of AI agents in seconds. With zero-config messaging gateways for WhatsApp/Telegram and a built-in Skill Market, we turn your single bot into a specialized workforce on enterprise-grade hardware. Stop debugging, start scaling.

MCPJam inspector is the local testing tool for ChatGPT apps and MCP servers. Build your apps locally with MCPJam's widget emulator and test against any LLM in the playground. Inspect your MCP server’s tools, resources, prompts, and OAuth flows. No more ngrok or ChatGPT subscription needed.

EaseLMS is an easy-to-install open-source Learning Management System for building and selling online courses: video lessons, quizzes, progress, certificates, analytics, and payments, with a separate marketing site and deep branding/SEO controls — self-hostable or available as a hosted service.

Voiden.md is an Open Source, offline-first, Git-native API workspace that unifies specs, docs, tests, and mocks using programmable Markdown blocks. Voiden helps developers, Testers and Technical Writers to build and Test APIs the way they actually work.

UI Layouts is a collection of 150+ production-ready UI blocks and components built for developers, designers, and teams. Ship faster, stay consistent, and deliver better DX without starting from scratch

You run Claude Code agents. You alt-tab 50 times a day to approve permissions. You lose your place. You miss prompts. Masko Code puts a mascot on your desktop that watches your agents in real time. Permission needed? Speech bubble. Press ⌘1. Done. Also: ⌘M jumps to the right terminal. Double-tap ⌘ to switch sessions. Tracks everything at a glance. Ships with Clippy and other mascots. Community adds new ones daily. Or generate your own at masko.ai. Free. Open source. MIT. 100% local.

ClawSecure is CrowdStrike for OpenClaw AI agents. 3-layer security audit, real-time Watchtower monitoring, agent marketplace and identity security, and full 10/10 OWASP ASI coverage. 41% of top skills are dangerous. 1 in 5 are sending your data to attackers. Secure your agents in 30 seconds for free. clawsecure.ai

screenpipe turns your computer into a personal AI that knows everything you've done. Record. Search. Automate. All local, all private, all yours. Your AI finally knows what you're doing

Human-proof your AI agents with this security skill suite. ClawSec is an open-source security skill suite created to harden OpenClaw agents against prompt injection, supply chain compromise, configuration drift, and unsafe runtime behavior. Purpose-built as a “skill-of-skills”, ClawSec wraps agents in a continuously verified security layer, validating what it runs, how it changes, and where the data is allowed to go.

LiveDemo is designed to empower founders, sales, and marketing professionals. With LiveDemo, you can effortlessly create captivating live demonstrations of your product, enabling you to: - Showcase your product's remarkable features to prospects. - Seamlessly capture hot leads by highlighting the value of your product. - Ultimately boost sales and drive conversions by letting your product speak for itself. LiveDemo is open-source on GitHub

InsForge is the backend built for agentic development. We offer everything AI agents need to build fullstack apps that scale. Our open source backend (2.3K stars on GitHub) provides databases, auth, storage, model gateway and edge functions accessible through a semantic layer that agents can understand, reason about, and operate end to end. Say the word, and you can deploy to InsForge Cloud or your own domain.

FratmScript is a fun compiler that lets you write JavaScript using Neapolitan-flavoured syntax and idioms — turning familiar coding into something that reads comme si stess’ parlanno ’nnapulitano. Perfect for devs who love creative coding, local culture, or just want to write code like a true Napoletano 🤌

Lumen is a Claude Code plugin that orchestrates 18 specialized AI agents across 6 PM workflows — PMF discovery, churn recovery, quarterly strategy, feature validation, GTM launch, and NRR analysis. Free to start. Open source MIT license. Built on Claude Agent SDK.

Desktop app for AI-powered video translation. Download videos from YouTube, transcribe with Whisper, translate to 30+ languages using GPT + Claude for quality review, and dub with AI voices. Free subtitle editing included. Open source & privacy-focused.

Automated code review agents that analyze pull requests, suggest improvements, catch bugs, and ensure code quality standards. Pick from 500+ models (Claude, GPT, Gemini, and several free options) to get instant feedback before merging.

Fast, scalable and cost-effective open-source observability platform. Monitor logs, metrics & traces with 140x lower storage costs than ElasticSearch. Start in 2 minutes.

AI coding agents edit your files blind; they can't see your running frontend. Domscribe closes the gap. Code → UI: Query any source location via MCP, get back live DOM, props, and state. No screenshots, no guessing. UI → Code: Click any element, describe what you want in plain English. Domscribe resolves the exact file:line:col and your agent edits it. Build-time stable IDs. React, Vue, Next.js, Nuxt. Vite, Webpack, Turbopack. Any coding agent. MIT licensed. Zero production impact.

Meteroid is a monetization platform built so early-stage teams can ship like they're already at scale: ✅ One engine for subscriptions, usage-based pricing, quote-to-cash, invoicing and business intelligence ✅ Support any Go-to-Market motion ✅ Open-source (not open core) and extendable. Self-host it or let us run it. Either way, it’s yours. 🚀

Flux records API executions so you can replay failures locally, fix them, and resume execution safely. Instead of guessing from logs, you get the exact request, inputs, and behavior. Same request. Same IO. Same outcome.

The simplest way to run OpenClaw. Atomic Bot works locally or in the cloud with your own LLM keys. It’s fully open source and free.

We have infrastructure as a code, network as a code but dont have anything as Risk As a Code. CRML is an open, declarative, engine-agnostic and Control / Attack framework–agnostic Cyber Risk Modeling Language. It provides a YAML/JSON format for describing cyber risk models, telemetry mappings, simulation pipelines, dependencies, and output requirements — without forcing you into a specific quantification method, simulation engine, or security-control / threat catalog.

Every morning you open 12 tabs, check the weather on your phone, wonder what time it is in Tokyo, forget which subscription renews today, and lose 50 minutes before your day even starts. Morgen replaces all of that with a single page. It's a local-first morning dashboard: weather, links, world clocks, site monitors, subscriptions, moon phase. It runs entirely on your machine. No account. No cloud. No tracking. You open it, you do your morning check-in, you close it.

Open source CLI turns your Markdown or MDX files into beautiful documentation. Bring your own AI model plus MCP support. Our hosted platform includes a visual editor, pending changes, custom domains, and team collaboration.

OpenClaw is powerful, but give it real credentials and you're exposed. Prompt injections steal API keys. Malicious skills grab passwords. IronClaw fixes this. Your credentials live in an encrypted vault inside a TEE — injected at the network boundary only for approved endpoints. The AI never sees the raw values. Every tool is Wasm-sandboxed. Outbound traffic is scanned for leaks. Built in Rust. Open source. Deploy on NEAR AI Cloud in one click.

A collective knowledge base where AI agents debug together via MCP. Ask questions, share fixes, and build collective intelligence.

A powerful CLI tool that writes, debugs, and ships code alongside you. Like Claude Code, but free and open source. AI-powered coding agent that analyzes tasks, creates actionable plans, and executes commands.

Most documentation tools live outside your repository. Your code changes. Your docs don’t. Drift is inevitable. Doclific addresses this problem in two ways: 1. Your docs all live in one place: your repo. This is the way it should be, ensuring engineers always have access to the correct version of docs. 2. Doclific features ERDs, architecture diagrams, smart snippets, etc. No more context switching. No more drift. Oh, and if you're lazy, you can just prompt the AI to generate docs for you!

Detour is the deferred deep linking tool built for developers, not marketers. It’s an open-source alternative to Branch or AppsFlyer – lighter and simpler. Free to start, easy to integrate, and built for teams who want full control over their linking infrastructure. Handle invitations, referrals, product/user links to create a seamless web-to-mobile flow and analyze traffic.

No single place shows your total Claude spend across tools. If you use Cursor, Claude Code CLI, Windsurf, Cline, or others — each burns tokens separately and you're flying blind. Claude Usage Tracker auto-detects 9+ tools, scans local session data, and shows everything in one dashboard: daily costs, model breakdowns, heatmaps, session logs, and monthly projections. Native macOS app or browser mode for other OS device. No cloud, no accounts, no telemetry. 100% free and open source (MIT).

Database tools haven't kept up. DBeaver is slow, pgAdmin feels stuck in 2010, and you're juggling 3 apps for 3 databases. QoreDB fixes this: one desktop app, 9 databases (PostgreSQL, MySQL, MongoDB, Redis, SQLite, DuckDB, SQL Server, CockroachDB, MariaDB), powered by Rust. Local-first : your credentials stay in an encrypted vault, your data never hits a cloud. Modern SQL editor, inline editing, SSH tunnels, production safety guards, full-text search. Core is Apache 2.0, free forever.

Sled lets you run your coding agent from your phone using voice. Coding agents need frequent input, but when you step away from your desk they just sit idle. Sled solves this by giving you a voice interface that connects securely to your local agent over Tailscale. Your code never leaves your machine. You talk, the agent works locally, and you hear the result read back. Works with Claude Code, OpenAI Codex, and Gemini CLI. Fully open source.

StackPatch is an open-source CLI that patches production-ready authentication directly into your codebase. Run one command and get Google & GitHub auth, protected routes, session management, Auth UI pages, and middleware & API routes all added as real files inside your repo -- No SaaS, No lock-in, no black box SDKs. Works with: Nextjs; React; Nuxt; Sveltekit; Astro

Build production-ready apps with everything you need to go live - authentication, database and monitoring included out of the box. Powered by an open-source framework designed for the AI era.

Ever shown up to an apartment only to find the "city view" is a brick wall? Window View lets you step inside any building on Google Earth and look out the window. Pick a floor, drop a window on any wall, and see exactly what's outside. A sun path overlay shows you when sunlight actually hits throughout the year. Free, open source, no account needed.

Our goal is simple but ambitious: to make deep-diving into code feel more like an adventure! Most of us know that Open Source is the backbone of the modern web, but finding the right project to learn from or contribute to can feel like searching for a needle in a digital haystack. Not anymore. We turn complex code repositories into interactive, AI-guided learning journeys. Choose how you learn best: Deep-dive, Socratic, Feynman, or traditional book format. No more 'wall of information' fatigue.

OpenClaw is the most popular open source AI agent on the planet. Running it yourself? That's the hard part. KiloClaw is a fully managed, hosted version of OpenClaw. We handle the infrastructure, security, updates, and monitoring so you can focus on what your agent actually does - not keeping it alive.

A tiny Rust menu bar app for macOS. Hold Fn to activate the mic, speak, release to paste. Audio streams to Deepgram Nova-3 in real time — no batch delay. Falls back to Groq Whisper. Open source, free.

The universal skill platform for AI coding agents. Auto-generate instructions with Primer, persist learnings with Memory, and distribute across Mesh networks. One CLI for Claude, Cursor, Windsurf, Copilot, and 28 more.

Ask questions about any repo and AI agents will clone, explore, and grep the source code to provide source-backed answers.

Angy is an open-source fleet manager and IDE for Claude Code, Anthropic API and Gemini API. Single-agent tools often generate code that fails at integration. Angy fixes this by orchestrating a deterministic multi-phase pipeline (Plan → Build → Test) featuring an adversarial Counterpart agent that strictly verifies all code. Using Git worktree isolation, multiple agents can build on your repo in parallel without branch conflicts.

Parallel Code is a macOS app that gives every AI coding agent its own git branch and worktree — automatically. Use Claude Code, Codex, and Gemini in parallel. Free and open source.

Arlopass is an open-source browser extension and developer SDK that lets any web app use your AI providers — Ollama, Claude, GPT, Bedrock — without ever touching your API keys. You approve each request. You pick the model. Your credentials never leave your device.

Axel helps you run AI agents and keep them fed. Queue up work, dispatch to the right agent, and approve or deny actions from one inbox. It's native macOS, keyboard-driven, and works with Claude, Codex, OpenCode, and Antigravity out of the box. Ship faster.

Kollect turns boring forms into real-time AI conversations. Users speak naturally, AI listens and dynamically guides the survey. You can even create forms by simply describing them.

A native macOS bookmark manager that attaches to any browser window as a sidebar. Open source, free, and local-first. Heavily inspired by Arc Browser's sidebar.

FastMCP 3.0 is a framework for building smarter AI context apps, not just tool servers. Pull tools and data from anywhere, reshape them, control access, track state, and run long tasks — with hot reload, versioning, and observability built in for production use.

Voquill is the open source alternative to WisprFlow. Type 4x faster by using your voice. It works in any app and runs natively on any operating system (MacOS, Windows, or Linux). Whether you're using agent mode or AI dictation mode, Voquill understands your intent and turns what you say into beautifully formatted text.

Workflow automation for everyone; just record your workflow once and run it on autopilot forever. Automated spins up its own browser to complete your workflow, no integrations required.

One order. Hundreds of AI agents plan, build, and learn together — sharing knowledge that compounds with every task. It looks like a tycoon game: pixel-art office, agents at desks, leveling up. But they do real work — CEO dispatches orders through the org tree, each role has scoped authority, and knowledge grows every session. Company-as-Code: roles and workflows in YAML/Markdown. Versionable, forkable — like IaC for organizations. npx tycono — one command. Local-first, BYOK, open-source.

Glance is an open-source MCP server that gives Claude Code a real Chromium browser with 30 tools. Navigate pages, take screenshots Claude can actually see, click buttons, fill forms, run multi-step E2E test scenarios, do visual regression testing, and record sessions all from your terminal. Built on Playwright with security profiles, rate limiting, and URL filtering. Battle-tested: 97% pass rate across 300+ test steps in production.

Building agents that call real APIs is painful. You end up hardcoding auth, juggling secrets in prompts, and writing glue code for every service. Jentic Mini is a self-hosted, open-source API execution layer that sits between your agent and the outside world. Your agent says what it wants to do. Jentic Mini finds the right API from a catalog of 10,000+, injects credentials at runtime, and brokers the request. Secrets never touch the agent.

I run an AI agent on OpenClaw that handles SEO tracking, content writing, code tasks, and more. Managing it through chat alone was getting chaotic. I needed a way to queue tasks visually, see what my agent was doing, track how much I was spending, and tweak its personality without editing files over SSH. VidClaw is that dashboard. It's opinionated, minimal, and built for people who actually run AI agents - not just talk about them. It's 100% self-hosted. Your data never leaves your server.

An open-source, independent instant messaging and VoIP platform. Built for friends, groups, and communities.

Browse, install, enable, and share AI agent skills across all major coding agents in one desktop app.

Cencurity is a security gateway that proxies LLM/agent traffic and detects / masks / blocks sensitive data and risky code patterns in requests and responses, while recording everything as Audit Logs.

Capture screenshots, annotate with powerful tools, and share instantly. The beautiful, free, and open source screenshot tool built for creators and developers.

OpenClix helps mobile teams run agent-driven local push campaigns without heavy setup. Create campaigns with smart triggers, suppression rules, and scheduling—then let agents analyze results, optimize engagement, and stop ineffective campaigns. Connect outcomes to retention metrics, ship faster, reduce notification fatigue, and iterate from real performance data.

Anthropic built Remote Control so you can manage Claude Code from your phone. We built the same thing for Codex. Message SuperTurtle on Telegram, by text or voice, and it runs Codex in a cloud sandbox. No laptop needed, no terminal open, just your ChatGPT subscription.

Drivebase is an open-source, cloud-agnostic file manager that lets you organize, upload, share, and collaborate on files across multiple providers from one interface. With Vault, its end-to-end encryption feature, your files are encrypted before they leave your device so only you can access them. Connect Google Drive, S3, Dropbox, or OneDrive, manage a unified folder structure, choose storage per file, and stay fully in control with a self-hosted, privacy-first setup free from vendor lock-in.

NothingHere is a macOS panic button. One hotkey, three things happen at once: 1. All windows disappear — every app gets hidden instantly 2. All sound goes silent — music, videos, everything mutes 3. A cover document opens — your pre-configured "real work" file Your screen goes from "definitely not working" to "hard at work" in milliseconds. Guard Mode sits in your menu bar — when armed, you're always one key press from a clean screen. Free, open source, ~5.9 MB. macOS 15.0+.

Baseline Core is an open-source skills system for AI agents. Run one command and your AI tools can research markets, write PRDs, plan sprints, and design user flows -- all grounded in your business context. 12 skills, 14 frameworks, 34 reference files. Works with Claude Code, Cursor, Codex, ChatGPT, Gemini, Windsurf, and GitHub Copilot. Free forever.

Lore is a lightweight "second brain" that lives in your system tray. Summon it with a keystroke to capture ideas, notes, or tasks instantly. Why Lore? 🛡️ 100% Private: Your data never leaves your machine. No API keys, no tracking. 🧠 Local AI: Powered by Ollama + LanceDB for secure, offline-first RAG. ⚡ Instant Recall: Ask questions in plain language and get answers from your own history. Own your memory. 100% local. Zero cloud.

The first dedicated open-source prompt library for Google Gemini. Finding professional prompts for complex tasks was a pain—formatting was messy and scattered everywhere. So I built this to fix it. We automatically scraped & cleaned 1,800+ prompts using a custom LLM pipeline (Qwen + Gemini) to ensure quality everyday. It's completely free, open-source, and so cool to use.

Claude Code deletes your session history after 30 days. Your work disappears. Unless you track it. CCgather preserves your journey. 🏆 Leaderboard | 💬 AI Community | 🎮 Levels | 📊 Heatmap Tokens ≠ skill—they represent exploration and passion.

AI memory API with 3 types: semantic (facts), episodic (events), and procedural (learned workflows). One API call extracts all three automatically. Killer feature: your agent completes a task → Mengram saves the steps → next time it already knows the optimal path with success/failure tracking. Works with Claude (MCP), LangChain, CrewAI, OpenClaw. Free, open-source, Apache 2.0.

Agent Skills is now an open standard, but each AI agent manages skills differently, making multi-agent workflows hard to maintain. flins acts as a dependency manager for agent skills, similar to npm. With commands like flins add, update, remove, list, and search, developers can manage skills consistently across multiple AI coding agents.

Octrafic is an open-source CLI for API testing. Point it at any OpenAPI spec or live endpoint, describe what you want to test in plain English, and let it handle the rest - from generating requests to validating responses and exporting a PDF report. No test scripts, no GUI, no mocks. Just a single binary. Works with OpenAI, Claude, Ollama, and any OpenAI-compatible provider

AI agents are powerful, but one wrong action could be catastrophic. Preloop is an agentic automation platform with built-in human approval layer. AI agents automate routine work across your systems, and when they attempt risky actions (deployments, refunds, data changes), Preloop intercepts and routes them for approval via mobile, Slack, or Teams before execution. You can use Preloop for automation only, approval gates only, or both together depending on your needs.

OpenBug is an open-source CLI that turns bug tickets into fixes. Paste a ticket, and the AI agent investigates your logs, reads your code, correlates across services, and delivers a diff. Every fix adds to a shared runbook in git — so your team gets smarter with every bug solved.

Prosaic is a minimalist writing environment for writers who want to focus on their craft, not their tools. Built for the terminal, Prosaic gives you everything you need and nothing you don't. Features: Markdown editor with live outline and word count. Focus mode strips away everything but your words. Daily writing metrics to track progress. Git-ready. Your files are plain markdown. No accounts, no costs, no cloud lock-in, no telemetry. Just you and your words. For macOS, Linux, and Windows.

Free, open-source macOS menu bar app that displays your Claude AI usage percentage. Get notifications before hitting your limit. Lightweight, private, and native.

Plano is delivery infrastructure for agentic apps. An AI-native proxy and dataplane that centralizes functionality that shouldn’t be bespoke in every codebase. For e.g. agent routing & orchestration, guardrail hooks, learning signals/traces, and smart routing APIs across LLMs. Use any AI framework, and ship agents to production faster. With Plano, developers stay focused on core product logic of agents. Product teams learn faster, and engineering teams standardize policies across agents.

Lightweight alternative to LaunchDarkly built with Bun & SQLite. Zero-latency local evaluation, percentage rollouts, and a sleek React dashboard. Deploy in seconds with Docker, Railway, or Zeabur. 100% Open Source. 🚀

Pangolin is the easiest identity-based VPN alternative. Securely connect to internal applications and infrastructure by granting users and groups access to specific resources. Reduce network and access control complexity. It's open-source, based on WireGuard, fast, and handles NAT traversal.

A terminal interface that translates plain English into shell commands. Stop memorizing flags. Just type what you want.

Parallax is a local-first AI orchestrator for software development tasks. It pulls work from Linear or GitHub Issues, creates isolated worktrees, generates a plan first, waits for approval, then executes changes and opens or updates the related branch and PR while keeping control on your machine.

OpenMolt lets you build programmatic AI agents in Node.js that think, plan, and act using tools, integrations, and memory — directly from your codebase.

Not another Claude Code GUI wrapper. claude-devtools doesn't run or modify Claude Code — it reads the raw session logs already on your machine and reconstructs everything the CLI hides. Every file path Read, every tool called, every diff Applied, every token consumed — structured into a visual timeline with per-turn context attribution, compaction visualization, subagent execution trees, and custom notification triggers. Works with every session you've ever run. Open source, runs locally.

Deploy, monitor, and scale your OpenClaw agents with a self-healing harness, fleet controls, and zero-SSH operations.

Mozzie is a local-first desktop tool that lets you orchestrate multiple AI coding agents in parallel. Create work items, spawn agents, and manage their execution from a single workspace. Each agent runs in its own terminal while Mozzie tracks tasks, context, and results. Mozzie works with existing coding agents like Codex, Claude Code, and Gemini CLI — it doesn’t replace them, it coordinates them. Open source on GitHub.

A startup compliance assistant that tracks obligations, fills government forms, and generates legal documents. Every obligation your startup owes can be discovered, tracked, and handled. Lexaclaw scans your documents to discover obligations, tracks every deadline, fills government forms, and generates legal documents from our templates. It also runs locally so your documents never leave your machine. Powered by OpenClaw. 100% free for startups. Just point your agent to our repo to get started

AI Memory is a privacy-first browser extension that captures, organizes, and lets you seamlessly recall your ChatGPT, Gemini, Claude, Perplexity (and more in the future!) conversations locally without data leaks.

A project built to help you transcribe, analayze and index your personal video library to help you search for the "exact" part of the video you're looking for with the help of local ML models and a local vector database. Your videos will never leave your computer or server (Docker support).

The AI-First Slide Authoring & Refinement Tool. AI Slide Creation Environment for Developers. Your AI agent builds the deck. You never leave the terminal. Create, revise, and share slides from the same CLI where you write code. You already use Claude Code, Codex, Gemini CLI, or Cursor for coding. Now ask the same agent to build your slide deck — without leaving your terminal. No app switching. No PowerPoint. Just flow.

Cntrl Bridge turns your Windows or macOS computer into a scriptable server. It runs a lightweight local API that lets you monitor system stats, control media playback, and manage power state from any device on your network. Built with Tauri & Rust.

VibePad is a free macOS menu bar app that maps your gamepad to keyboard shortcuts. Built for AI coding tools like Claude Code and Codex where the workflow is mostly Accept, Reject, Scroll, Navigate and Dictate. X - approve, O - reject, L2 trigger = hold-to-talk dictation, Right stick = scroll. Config is a JSON file if you want to remap anything. Native Swift. Open source. No account, no subscription, no catch. The idea started as a joke - then I built most of VibePad using VibePad itself.

Free, open-source desktop app that auto-detects every AI client on your machine — Claude Desktop, Cursor, Windsurf, VS Code and more. Manage all your MCP server configs in one place, install from the marketplace, and back up with one click. macOS, Windows & Linux.

OpenFlowKit is a 100% free, open-source diagram engine for engineers, architects, and product teams who care about craft.

Formerly Moltbot and Clawbot, OpenClaw turns your computer into a 24/7 personal agent accessible from any chat app. Control your browser, execute shell commands, manage files, and automate workflows via WhatsApp or Telegram. Features persistent memory, full system access, local privacy, and 50+ integrations.

Moltcraft is an open-source isometric pixel dashboard for Moltbot. Instead of reading terminal logs, watch your AI agents walk around a living world. Each agent becomes a pixel character. Click buildings for real data: cron jobs, token usage, skills, channels. Talk to any agent with voice. Zero npm dependencies, pure HTML/CSS/JS, ~2MB, runs on a Raspberry Pi. One command: npx @ask-mojo/moltcraft. Cloud version coming soon — join the waitlist at moltcraft.xyz.

NetNewsWire is a free and open source RSS reader. It's fast, stable, and accessible. NetNewsWire shows you articles from your favorite blogs and news sites - and keeps track of what you've read. If you've been going from page to page in your browser looking for new articles to read, let NetNewsWire bring them to you instead. It's easy and fun - and it puts you in control of your news.

Vite+ is a new unified toolchain and entry point to web application development. Manage your runtime, package manager, and frontend stack with one tool. curl -fsSL https://vite.plus | bash

A native macOS desktop client for Hacker News that goes beyond the website. Browse stories in a visual grid with article thumbnails, read articles side-by-side with comment threads, and use reader mode to focus on content. Built entirely with SwiftUI for a fast, native experience with full dark mode support, 15+ keyboard shortcuts, built-in ad blocking, and adjustable text scaling. Log in with your HN account to bookmark stories, hide items, and sync across sessions. Free and open-source.

MacMonitor is a free, open-source system monitor built specifically for Apple Silicon (M1–M4). It lives in your menu bar, updates every 2 seconds, and opens a full dark-mode dashboard showing CPU per-core usage, GPU, memory, battery health, power rails, network, disk I/O, and your top processes — all from the metal, no subscriptions. Install in one command: brew tap ryyansafar/macmonitor && brew install --cask macmonitor

Images and videos are bigger than they should be. They eat disk, bandwidth, and slow everything down. We tried the usual mix of tools, flags, and formats. It worked… sometimes. We just wanted something consistent. So we built Optimo: a CLI that makes media smaller, every time. Built on ImageMagick + FFmpeg. Free and open source.

Modern instant bug reporting platform. An open-source alternative to jam.dev and marker.io.

Turn your messy desktop into a Tinder-like game. A native macOS app that lets you swipe left to trash and right to keep. It detects duplicates, groups similar files, and makes cleaning up screenshot_final_v2.png actually satisfying. 100% Free & Open Source.

Echolon - A powerful, local-first API client. Open source alternative to Postman with Git integration, offline support, and multi-protocol capabilities.

Open-source Agent OS built in Rust. 7 autonomous Hands that work for you on schedules. 16 security systems. 53 tools. 40 channels. 27 LLM providers. WASM sandbox, Merkle audit trail, taint tracking. Single binary.

Cohere Transcribe is a state-of-the-art, 2B open-weights speech recognition model. Optimized for enterprise workloads, it delivers high throughput and a leading 5.42% WER across 14 languages, making it ideal for private, local, or desktop deployment.

Google Workspace CLI lets humans and AI agents control Drive, Gmail, Calendar, Sheets, Docs, and more from one CLI. Built from Google’s Discovery Service, it stays up to date automatically and includes 100+ agent skills to automate workflows without the MCP context tax.

DialogLab is a research prototype that provides a unified interface to configure conversational scenes, define agent personas, manage group structures, specify turn-taking rules, and orchestrate transitions between scripted narratives and improvisation. Designers can 1) configure group, party, snippet characteristics, 2) test with simulation and live interaction, and 3) gain insights with timeline view and post-hoc analytics.

MiniMax M2.7 is a self-evolving AI model that helped build its own capabilities. It can create agent harnesses, collaborate via Agent Teams, and handle complex tasks like coding, debugging, and research. With strong SWE-Pro performance and reduced intervention time, it moves beyond static AI into systems that continuously learn, adapt, and execute complex work with minimal human input. Available via API and MiniMax Agent for builders pushing AI-native workflows.

Breadcrumb is the Plausible of LLM tracing. Self-hosted, open source, and built for developers who just want to understand what their agents are actually doing without the enterprise bloat of LangFuse or LangSmith. Three lines to get your app traced. An LLM watches every trace and automatically flags issues: wrong tool calls, looping agents, oversized models and cost spikes, all before you even know something's wrong. Ask questions about your traces in plain English and get charts back.

Qwen just released the Qwen3.5 Small Model Series — 0.8B, 2B, 4B and 9B. Native multimodal with improved architecture and scaled RL. 0.8B and 2B are tiny and fast for edge devices, 4B makes a strong lightweight agent base, and 9B is already closing the gap with much larger models. Base versions released too.

TranslateGemma is a new suite of open AI translation models built on Google’s Gemma 3. It enables high-quality communication across 55 languages, combining strong accuracy with exceptional efficiency. Designed to run on mobile, local devices, and cloud environments without compromising performance.

FreeCAD 1.1 is a massive update to the highly capable, free, and open-source 3D CAD/CAM/FEA modeler. It introduces major quality-of-life improvements including transparent previews, interactive draggers, new CAM tools, and enhanced assembly features.

FasterGH is a fast GitHub mirror UI backed by Convex as a real-time cache/sync layer. GitHub is the source of truth; Convex is the low-latency read model.

Trusted by 5M+ developers, the Cline CLI brings autonomous coding directly to your command line. Fully open source, redesigned from the ground up. Features parallel agents, headless mode for CI/CD, and ACP support for any editor (Zed, Neovim).

Building AI features today means navigating a maze of competing frameworks, juggling provider API keys, and wiring together components that were never designed to work as a system. Most engineers spend weeks on infrastructure before writing a single line of application logic. RocketRide is an open source, developer-native AI pipeline tool that changes that. Pipelines are first-class JSON objects. Models are interchangeable. The entire workflow lives in your IDE.

OpenYak is an open-source desktop AI agent that handles your files, documents, and workflows — locally. 100+ cloud models, 20+ built-in tools, Ollama for offline use. Your data never leaves your machine. Free to start.

Package Mate is an open source, native Apple Silicon manager built in Go to fix the friction of macOS dev setup. This tool offers a seamless fire-and-forget workflow by allowing you to queue Homebrew installations in the background through automated daemonization while providing a high-performance, zero dependency interactive dashboard for managing your tool stack. It leverages deep version intelligence to resolve conflicts with unmanaged binaries and mirrors setups via base64 strings.

Free macOS app for running multiple Claude Code sessions side by side. Split panes, 7 themes, command palette, agent kanban board, AI memory, prompt compression, usage tracking, and broadcast mode. Native Swift/AppKit. Signed and notarized.

Full observability inside NVIDIA NemoClaw sandboxes. One command on the host, every sandbox gets covered. See every thought, tool call, and token cost in real time. Brain activity, flow visualization, memory monitoring. All E2E encrypted. 95K+ installs. 100+ countries. Open source (MIT). Cloud sync: $5/sandbox/month.

Remodex is an open-source iOS app that remotely controls Codex running on your Mac. From your iPhone: create threads, run subagents, push git commits, use skills and /commands and more Codex stays on your Mac. End-to-end encrypted, the relay never sees your prompts. Pair once with a QR code, reconnects automatically. Open source. Self-host the relay if you want full control.

Open source anti-detect browser. Create unlimited local profiles with unique fingerprints, proxy support, and full browser automation.

Free open-source macOS menu bar app for Claude Code. Browse full session history, search across conversations, track token costs per project with Anthropic or Vertex AI pricing. Built-in secret detection scans every session for leaked API keys, tokens, and credentials. Config health linter runs 19 rules against your CLAUDE.md files, skills, and hooks. Works with Enterprise API deployments (no cookies needed). Native Swift/SwiftUI, 100% local, zero telemetry. MIT licensed.

A lightweight Mac app cleaner that lives in your MacBook’s notch. Drag an app to the notch, drop it, and Ditch removes it along with leftover caches, logs, and preferences. Fast, safe, and fully open source, no bloated installers or complicated menus, just effortless Mac cleanup. Key Features: 🪶 Lives in your MacBook notch - always accessible 🧹 Deep cleanup: caches, prefs, logs, containers 👀 Preview before deleting 🗑️ Everything goes safely to Trash ⚡ Lightweight and lightning-fast

OpenBerth is a self-hosted deployment platform that AI assistants can use natively. Tell Claude, your agent, or any AI tools you use on your phone or desktop to deploy an app, it does. Store and use secrets securely, AI references them by name without seeing values. Works from CLI, AI chat, or a single file drop. Your server, your rules. Connect any AI tools to MCP server and start building.

The fastest way to deploy Docker containers to your own server. One command. Automatic HTTPS. Zero config. Platforms like Vercel and Railway have great DX but are expensive and lock you in. Self-hosting is cheap and flexible, but getting containers live with HTTPS is a pain. zero closes that gap — deploy any Docker image to your own server with a single command. Auto HTTPS, zero-downtime deploys, health checks, rollbacks, and preview environments. No YAML. No web UI. Just a CLI that works.

Gemma 4 is Google DeepMind’s most capable open model family, delivering advanced reasoning, multimodal processing, and agentic workflows. Optimized for everything from mobile devices to GPUs, it enables developers to build powerful AI apps efficiently with high performance and low compute overhead.