Superset for iPhone
Control coding agents on your computer from your pocket. Review and ship PRs. Claude Code, Codex, other terminal agents.
Feature a build or sponsor the directory. DM @pranavmore69
Systems that take a goal and come back with finished work: mailers, reviewers, traders, on-call bots. Harnesses over chat UIs. 132 builds on this page, each with the original post.
Control coding agents on your computer from your pocket. Review and ship PRs. Claude Code, Codex, other terminal agents.
Meta's personal AI agent. Own Secure VM, own browser, does the work across apps. Powered by Muse Spark 1.3.
Personal agent on your computer. Apps, files, calendar, notes, messages. You pick what it can touch.
Developer connectors for Muse. You bring the API. The agent, browser, and user context come with it.
Native Mac clicker for Claude Code and Codex. Jev decides, OmniParser and Apple Vision stay on-device. Apple silicon, BYO token.
TypeSafe's System One model. Typed decisions, no chat. Trained with RLCD. Output tokens free.
Production layer for Jev: policy, confidence gate, shadow mode, recipes, eval CLI.
Native macOS GUI for the Pi coding agent. SwiftUI, projects and sessions, diffs beside the chat. MIT.
Free agent skills that force a straight verdict, plain steps, and a next action. Built by a non-engineer running a software company.
Native Mac app for Vercel's fx coding agent. Draws every pixel on the GPU through GPUIX. OSS.
HTLC and PTLC deal-making for agents that meet in a chat room. Spec, library, and a walkthrough. First-party from Flop Labs.
Anthropic's public Mythos-class model. Long-horizon coding and research, cheaper cache reads. Live everywhere today.
Official MIT harness from DeepSeek. Everything is a plugin: models, tools, sandboxes, loops, and UI. Run with npx @deepseek-ai/dsh web.
One Docker box, one API. Codex, Claude Code, Hermes, Pi, and DeepSeek Harness on your machine. Apache 2.0 plus an open spec (UHP).
OpenAI's open Codex harness as an embeddable agent loop. Your app owns the UI, tools, and approvals. SDK, exec, or app-server.
An agent harness written in pure Bash, now with a web UI. First-party from Di Zhang.
Ask Codex or Claude Code to add VRM facial expressions. First-party, with a demo of the avatar moving.
Claude Code-style workflows that run in any harness with any model. A skill, not another CLI.
DeepMind's harness for autonomous research runs that last days to weeks. Used internally on Simply, now public.
Inference runtime, local API, agent loop, and training provenance for Dot's 14B model. Company open-sourced the engineering layer.
Teknium's stand-alone Hermes Agent plugin. Search the web as it was on a given date, like a Wayback Machine for agents.
An agent skill that picks a Nerd Fonts glyph by name in React, React Native, or Expo. Over 10,000 icons.
A Mac Mini hosting agents calls you on FaceTime Audio. Open-source adapter. Use a dedicated iCloud account on the box.
Desktop app that runs Claude Code, Codex, Cursor, and 26 other coding-agent harnesses in one place.
Tiny agent-friendly CLI wrapping Paper MCP so it works with delta.dev.
Sandboxed AI harness in Zig. Landlock, eBPF, and namespaces isolate tool calls. WASM plugins over IPC.
Internal agent apps Rome runs the company on, open-sourced so you can see how they actually work.
14-agent security auditing harness in the terminal, rebuilt for Web3 on OpenCode.
RAG pipeline that ingests the official XRPL docs so an agent can build against them.
Turn a repo into a knowledge graph and hand it to Codex, Cursor, or Claude Code over MCP. Runs in the browser.
The Apple Notes CLI Apple never shipped. Reverse-engineered, not an AppleScript wrapper. Tags, PencilKit drawings, Markdown export.
Tag an AI in Slack the way Claude Tag works, but pick any model. Live with 10 teams.
Shopping agent inside iMessage. Photo or text in, Shopify catalog search, one cart per merchant, Shop identity with a budget. No card details leave the phone.
USB webcam capture for agents, now with a rotation protocol so a claw can look around a 360 camera.
Bureaucracy for coding agents. They file requirements, get designs confirmed, claim tasks, and leave a paper trail on every merge.
vLLM plugin that fixes compaction and thinking-mode bugs when running Qwen 3.8 27B as a local Hermes agent.
Rust SMTP server with an API, built because they wanted agent inboxes. Linux, Windows, Docker. Now open source.
Thin harness so an agent can drive a full Omarchy desktop: Hyprland, AT-SPI, real browser, files.
Agent skill that turns any video into a scroll-controlled website.
Self-hosted knowledge vault for Hermes: ingest articles, YouTube, RSS, bookmarks, then search them from the agent.
Agent skill that teaches Claude how Cisco pyATS works, without standing up an MCP.
Open-source AI IDE for Claude Code, Codex, Grok, Cursor, OpenCode. Parallel agents, each in its own worktree.
Claude Code drives a real Android phone over MCP and Shizuku. No root, no model on the device.
Harness that gives every AI bot its own directory, browser profile, cookie jar, and environment. Grants expire; the door closes.
Create and manage Grok Bot agents and groups from your terminal. Send prompts, inspect threads.
Turns each agent rule into a function that checks the agent's work, instead of hoping it follows the prompt.
Profile-aware achievement plugin for Hermes Agent. See which agent unlocked what. MIT.
Poole Adaptive Reasoning Stack as a skill: generate exact binaries, research, and audit with PARS.
Open-sourced agent memory. Mid-session compression cut 61% of token usage in their case.
Open agent memory on Cloudflare Workers and Durable Objects. Profiles, extract, recall. One-click deploy. MIT.
Lauren's Cursor plugin for rigorous engineering and verification: /goal, /loop, /swarm, Full Autopilot. Public as /poteto-mode. She shipped 1000 PRs last month on it, including Grok Bot.
OpenRouter for tools: 2,600 agent-friendly APIs, pay per call, 0% markup, search by task. Open source.
Open-source MCP sidecar: connect trusted clients to Hermes Agent without modifying Hermes. Files, memory, skills, cron, Codex.
YC-backed open-source computer-use alternative to Claude Cowork.
Open-source GUI on the real Pi runtime. Alpha.
Consultancy OS: reads email, remembers the client, runs loops, hosts apps. Open-sourced.
Universal model router for AI coding agents. 11k GitHub stars.
Persistent local memory for coding agents. One Python file, no dependencies, explainable recall.
Agents get limited USDC spend via MetaMask permission, GenLayer policy, 1Shot execution. No custody.
Svelte agentation site.
Minimal Pi extension so agents can /grill-me you. Unlimited questions.
Pixel office for coding agents on your Mac: Claude Code, Gemini CLI, Cursor and Codex each get a desk.
Anticipatory co-pilot skill pack for Cursor, Claude Code, Antigravity and more. Built with Gemini 3.7.
Hermes builder tools updated after running Qwen 3.8 27B so smaller local models can break work into parts.
SpaceXAI's Eric Zakariasson lists 100 real Grok Bot jobs: overnight outbound, CRM rebuilds, Gong win/loss, Salesforce from a phone, security questionnaires.
Run Grok Bot locally with a swap-in model backend (Codex OAuth or any OpenAI-compatible API) and a local computer desktop you can take over over noVNC.
Open-source Grok Bot alternative: persistent bots with memory, routines, a shared Linux desktop, and the ability to spawn more bots. Bring your own model.
Open-source Claude agent definitions that pretty much fully automate a security audit.
oh-my-pi extension for best-of-N + self-verify: sample several solutions, rank them with the same model as verifier.
Shared local brain for Claude Code, Cursor, Codex and OpenCode. MCP plus CLI so agents stop redoing each other's work.
Personal search over X, YouTube, RSS, Bluesky and Mastodon. Web UI plus MCP so agents can query it.
A 24/7 agent for solo game devs. Watches Steam, itch.io and Discord, replies to players, files bug tickets and drafts weekly patch notes.
An AI-native Layer 1 experiment on the Canopy testnet, giving autonomous agents their own infrastructure and economy.
Open-source AI agent that reviews your code and auto-fixes it.
MIT tmux pane that renders what your coding agent can see and you can't: images, video, PDFs, diffs and X cards. Bash 3.2, nothing leaves the machine.
MCP bridge so ChatGPT can operate your real Mac: shell, files, PTYs, background jobs and read-only Codex history. MIT. No model calls.
AI agents that crawl the App Store and Play Store around the clock and deliver MASVS and GDPR digests the moment a new app ships.
Open-source AI agents that run a full pentest end to end, from recon through exploit, reverse shell and pivot, then write a client-ready PDF report.
An AI cowork harness tested heavily against open-weight models. Runs well on DeepSeek V4 Flash and Meta’s Muse Glimmer 30B, which is agentic out of the box.
Two cancer tools from a practicing oncologist: an AI tumor board for clinicians, and a multi-agent patient-facing tool that answers the questions doctors have no time for.
An agent built with Claude, GitHub and Notion that finds recently merged PRs, summarizes what shipped, and posts a fresh digest to Notion every Monday.
Generates software installation instructions for application owners and packaging teams, cutting the manual documentation work.
The open-source Grok Bot clone now schedules agents on a real calendar instead of a typical routines UI.
macOS / Mac mini scheduler for boring tasks and headless agents: research runs, scripts, software updates, programmatic SEO.
Free, offline, open-source workpaper binder for accountants. Agents connect over local MCP, every change is journaled, and humans still sign off.
Agent tools for any SDK that live-refresh and can be searched or described, plus a durable agent mailer along the lines of Oban for JS.
Free open-source Postgres intelligence for agents and apps. Finds unused indexes, slow queries and drift. Written in Go. No dashboards.
Skill that lets your Claude agent get a real payment card (US or international) and spend via x402 + USDC.
AI teammates from Cursor + SpaceXAI that sign into your tools, do real work, and return finished results. Includes 'Teach A Task' feature that watches your screen and turns workflows into skills.
Fully shipped AI agent with persistent memory built in less than a day.
Chat interface that lets you trade across five different chains from one conversation.
Open-source meta-harness for building, running and observing AI agent workflows.
Open-source coding agent written in Rust with desktop + CLI interfaces and self-hosted remote access.
Lightweight local control plane and state kernel for long-running AI agent loops with goal tracking and auto-wake.
Open-source CLI agent for authorized pentests. Plan, act, observe, verify, report. Local or hosted LLMs. Burp + MCP.
Persistent Grok-based AI agent with its own wallet, identity and ongoing lore around the “Year of the Singularity”.
Practical system of 47 working agents built with n8n and Claude, including the three prompts that actually work.
Connect Claude-style agent workflow directly to physical robots. Describe the task and the system writes the code to execute it in ~20 minutes.
Open-source agent, editor and security-review tool. Headless CLI, any OpenAI-compatible endpoint.
Open-source extension that lets Pi sessions talk to Claude Code sessions in real time (agent-to-agent messaging).
Open-source tool that gives coding agents a persistent map of your codebase, cutting token burn by 30 to 99.7% across Claude Code, Codex, Cursor and more.
Free open-source Claude Code plugin that lets your parallel sessions see what each other are doing, built on the new cross-session messaging feature.
Open-source repo that grew from an 11-agent experiment into 18 reasoning personas with multi-provider routing and confidence-weighted verdicts.
Fully open-source agent operating system that runs on mobile, desktop and web with voice + chat control.
AI that has its own phone number and email. Makes calls, waits on hold, and books follow-ups. #1 on Product Hunt.
Physical robot you can program directly with Claude Code or Codex to help with real-world tasks like working on cars.
Coding agent you speak to instead of typing. Say what you want and it ships in the background.
Agent that schedules meetings and books reservations 100% over text message. No app required.
Cloud hosting for agents starting at $3/month. Choose harness, model and capabilities.
Upgraded Mac notch app that now pings your iPhone and taps your Apple Watch when Claude Code, Cursor or other agents finish or need input.
AI tool focused on meeting intelligence and action extraction.
Voice agent platform for building conversational AI experiences.
Command-line AI agent that works directly with Gmail.
AI agent that risk-scores pull requests and can automatically approve low-risk ones.
Open-source self-improving coding harness where sub-agents are function calls in a persistent kernel. Scored 95.5% on ARC-AGI-3, above the human expert baseline.
Kenton Varda's remake of his startup Sandstorm as an AI agent workspace with connectors, built entirely on Cloudflare Workers. #1 on Product Hunt.
Rust open-source parser that converts PDF, DOCX, PPTX and 10+ formats to clean Markdown in under 5ms, built for agents. 12.5k stars in days.
40 interconnected AI systems that run a real 7-figure company with almost zero human decisions.
Fully automated content system using Claude that writes posts, replies to comments, tracks engagement and repurposes content.
AI coding agent that takes a single prompt and ships a complete project.
Open-source control plane for AI agents. Defense by default. Used by thousands of devs a month.
Operating system style environment for running and managing AI agents.
Agent framework designed to automate migration of enterprises to open SaaS running on ICP cloud engines.
Fully open-source agentic CRM built as a modern alternative to expensive SaaS tools. MIT licensed.
Open-source tool that gives frontier models effective 100M–1B token context by using a graph-based retrieval system.
Fully playable Three.js FPS built by Claude Opus 5 from a single prompt using a multi-agent builder and critic loop. Prompt and repo open-sourced.
Open-source desktop AI coworker that delivers finished work: polished documents, Slack messages and calendar updates, with 35 connectors and permission tiers.
Run Claude Code, Cursor, Grok, Codex, Antigravity and more side-by-side inside the Mac notch with smart notifications.
One CLI that gives any AI agent read and search access to X, Reddit, YouTube, GitHub and more with zero API fees. 65k+ stars in about a month.
Platform that lets you build apps and agents with a single prompt.
Autonomous AI agents running on the Arc network.
Drop-in payment modal that lets any agent or endpoint accept micropayments using the x402 protocol.
Introducing Superset for iPhone. Remotely control agents running on your computer. Review and ship PRs from your pocket.
Muse for Mac is out today! It works across apps, files, calendar, notes, and messages on your computer. You control what it can access. The team is shipping fast. Download at ai.meta.com/muse/download
Opening access for developers to build Muse connectors. You bring the API -- Muse brings the agent, the browser, and the context of what the person actually wants. People reach your service just by asking for it, and their agent takes it from there. New connectors are live Show more
Give Claude Code or Codex a native Mac clicker. I open-sourced Jev macOS Loop. The README has a prompt your agent can follow through setup and verification. Apple silicon. BYO OpenRouter, Vercel or TypesafeAI token. github.com/jcpsimmons/jev…
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x Show more
TypeSafe Jev is great at structured decisions. I open-sourced the missing production layer: jev-harness → policy → confidence gate → shadow mode → recipes → eval CLI github.com/AntonioCoppe/j…
This is just mad 🤣 I built this @pidotdev native macOS client, ChatGPT style: - projects / sessions / pinned - supported packages / file view - 3rd party provider settings Code (MIT) + macOS app available at github.com/hieunc229/Pie
I am not an engineer. I run a software company and use coding agents daily. Their output is written for engineers, so I built Open Steps. Free skills that make an agent give a straight verdict, plain steps and a real next action. 250+ stars in a month. github.com/kharmanskyi/op…
I made a Mac app for fx, the coding agent from Vercel. fx-ui runs libfx inside the app and draws every pixel on the GPU through GPUIX. Bring a Grok or Codex subscription, or a Vercel AI Gateway key. It's open source: github.com/zaidmukaddam/f…
Two agents meet in a chat room. One wants work done, the other wants paying. Neither can afford to go first. The old answer is a hash lock and a deadline. tclk/1 runs one over a room both agents can already reach. github.com/flop-labs/tclk
We’re introducing Claude Fable 5.1 and Claude Mythos 5.1. They're the world’s most advanced models for coding and knowledge work.
🧩 DeepSeek Harness v0.1 is now available in Developer Preview! 🔹 We’re opening it up to developers building agent harnesses worldwide and open-sourcing the codebase in MIT license. 🔹 Powered by the Cordis meta-framework, DeepSeek Harness is an agent harness built around one Show more
The world's first unified interface for agent harnesses is now open source. Apache 2.0. No more harness engineering for your product. Sandboxes, sessions, permissions, tools, context, artifacts: that whole layer is on us, and it is open source now. This weekend, your time goes Show more
Teams are using the open-source Codex harness to bring agents into the tools they already use, from internal apps to operations dashboards. Their applications control the interface, context, tools, and approvals while the harness handles the agent loop. developers.openai.com/blog/codex-as-…
We have built a Agent harness with pure Bash script! Now it have webui support! github.com/trotsky1997/har
イラスト1枚でVTuber3Dモデルが生成できる「VTuberオリキャラメーカー」が話題ですね! 3Dモデルと言えば、CodexやClaude Codeに依頼するだけでVRM感情表現を拡張出来る『VRM Expression Agent Harness』というツールを公開しています アバター表現を拡張したい方はぜひ github.com/shinshin86/vrm…
Claude Code workflows, but they work in any harness with any model, check my workgraph skill :) github.com/av/skills/tree…
Glad that others are finding Amplio helpful. It is our main harness for autonomous long research runs spanning days to weeks. It is heavily used by our team for automated AI research / RSI on top of the Simply codebase. Open sourced at: github.com/google-deepmin…
Harness matters. We test Gemini 3.1 Pro with amplio, a harness for robust and long-horizon runs and super fit for auto-research. We then get much better results than with Gemini CLI. Harness: github.com/google-deepmin… FrontierCS is an open-ended benchmark for Computer Science
We’ve open sourced the engineering layer behind Dot Reflex 14B. The repo includes the inference runtime, local API, trajectory schema, agent loop examples, tests, benchmark receipts and training provenance. GitHub: github.com/usedotai/dot-r… Model weights: Show more
Made a stand-alone Hermes Agent plugin for BackSearch - a wayback machine-like SaaS for agents - adds two tools, and requires their API key. Let me know if you like it github.com/NousResearch/h…
🔍 Introducing BackSearch. LLMs are increasingly asked to predict the future, but a good backtest requires a snapshot of the internet at a point in time. BackSearch allows LLMs to search the web as it was on a particular date. It’s great for: 🔮 Forecasting and prediction
🎉 I built an agent skill so Claude Code, Codex, or Cursor can pick a Nerd Fonts glyph by name while you work in a React, React Native, Expo, etc... More than 10,000 icons for free! github.com/massimodeluisa… > npx skills add massimodeluisa/nerdfonts-skill
I open sourced the facetime adapter for agents! Make sure to use a new iCloud account dedicated to your mac device that is hosting your agents: github.com/kingbootoshi/f…
We’ve built exactly this :) AO lets you manage Claude Code, Codex, Cursor, and 26 coding-agent harnesses from one desktop app. aoagents.dev github.com/Untrivial-ai/a…
delta.dev did not support mcp yet, so i had do make a workaround to get paper working. Made an little agent freidnly cli thats just a wrapper around the paper mcp. try it if you use delta. github.com/lassejlv/paper…
Just released a sandboxed AI harness. It's written purely in Zig and utilizes things such as landlock, eBPF, and namespaces to isolate tool calls for LLM's. It also utilizes WASM for plugins in an isolated process and fed back to the session over IPC. github.com/Midstall/chock
We’re not just building @RomeAILab — we’re running our company on it. As an AI-native company, we use Rome every day to help us get work done. We’ve open-sourced some of the apps we use internally so you can see how we actually work with our agents. github.com/rome-os/rome-a…
🚨 Introducing AuditCode v1.0. An autonomous 14-agent security auditing harness for your terminal. Built on top of @opencode + PentestCode. Rebuilt for Web3 by @itsdarktoday & @0xscarfac3. Try it now. 🦀⚡ 🔗 github.com/itsdarktoday/a…
Are you using an AI agent to build on the #XRPL? I built a RAG pipeline for ingesting the official docs. Future updates will include the various libraries. github.com/joshuahamsa/xr…
GitNexus MCP plus CLI in one package is how I want agents to talk to a repo github.com/abhigyanpatwar…
I built the ultimate Apple Notes CLI. Apple never made a real API for Notes. So, I reverse-engineered one. NotesCTL covers ALL the features of the Notes app. NotesCTL can: - Create notes with support for every element. Tags, rich links, attachments, checklists, headings, Show more
You might've heard of @OpenAI, @OpenRouter, @opencode We're excited to launch @TryOpenTag : the model-agnostic Claude Tag. If you've seen Claude Tag, you know the interface. Tag an AI in Slack, hand it a task, get the result in the thread. We think that's the right interface. Show more
I built a shopping agent inside iMessage. Open sourcing the code: github.com/gil--/ucp-agen… 1. Send a msg or photo. 2. It searches Shopify’s global catalog. 3. Builds one cart per merchant. 4. Connects Shop identity w/ budget. 5. No cc details exposed. Why Shopify? Shop identity Show more
Instinct shows how far LLM computer use has come but it’s still awfully slow 😴 and no one wants to share payment (or address) credentials with a 3p. 😱 Here’s a short demo of my agent, Concierge using Shopify’s UCP tools and our Shop identity tokens to directly checkout with
I added the rotation USB protocol to github.com/steipete/camsn… and told my claw to look around- now it's having fun with my 360 webcam.
I built a bureaucracy for my coding agents. They now have to file requirements, get designs confirmed, claim tasks, and leave a paper trail on every merge. Somehow it’s less chaotic this way. github.com/kidus-tiliksew…
Have been using qwen3.8:27b for my local hermes agent. Had some issues with compaction and thinking modes. This plugin I made fixed all the issues I was having, maybe others will benefit too github.com/benthecarman/h…
Last year we replaced our entire email infra because we wanted agent inboxes and better send handling across our platforms. We've now open-sourced Sentio, a Rust-based SMTP server with an extensive API. Linux, Windows, Docker. Contribute: github.com/truespar/sentio
Perfect timing with the “malleable computer for the age of agents” vision. I built a thin harness that lets an agent fully control an Omarchy desktop (Hyprland, AT-SPI, real browser, etc.): github.com/fabiopauli/oma…
too many people DMd me about this so i open sourced it if you read a lot of stuff and use hermes and wanna super charge it, go for it github.com/ianlapham/supe…
I made a pyATS Skill for your agents! github.com/automateyourne… I love my pyATS MCP but this is a great tool-free alternative - just teach Claude about how pyATS works and away you go
Today we're launching Proliferate, open source Codex for working with any agent- here's how we use Proliferate to build Proliferate.
Claude Code just opened Xiaohongshu on my actual Android phone.ʚ/•᷅•᷄\ɞ The path is: Claude Code → MCP → HTTPS Bridge → Android Agent → Shizuku → 📱 No root, no AI credentials on the phone, no model polling. I open-sourced it: github.com/2005selene2005… Demo soon 👀
I built a harness that gives every AI bot its own directory, its own browser profile, its own cookie jar, its own environment. -> github.com/Archive228/cub… Bot asks for a service → a grant names it, or it's denied. Grant expires → the door closes on its own. Handoff → only the Show more
Just shipped grok-bot-cli: create and manage Grok Bot agents and groups from your terminal, send prompts, and inspect threads. npm i -g grok-bot-cli github.com/ScriptedAlchem…
Frustrated your coding agent keeps ignoring your rules? What if every rule could take on a life of its own? I built RulesAsPrograms which turns each rule into a function that checks agent's work. Instead of hoping agent follows rules, rules check agent github.com/programasweigh…
Open-sourced: Profile-Aware Achievement plugin for Hermes Agent. See which agent have unlocked what achievement. MIT: github.com/BkashJEE/herme…
Your Hermes Agent finally has a skill tree. Hermes Desktop can now show which achievements each agent has unlocked—and how many. Your profile and SOUL.md define who the agent should become. Achievements reveal what it has demonstrated so far. Use them to: → Identify
I made PARS//Poole Adaptive Reasoning Stack into a skill! Generate exact binaries, research, and audit with PARS. Currently only tested and used with Codex 5.6 Sol. Get the skill.md here: github.com/rookepoole/PARS Support my research and builds here: buymeacoffee.com/rookepoole
Agreed, been working on this. Agents re-learning who you are every session is the real waste. Mid-session compression alone cut 61% of token usage in our case. open-sourced the whole thing 👉 github.com/TencentCloud/T…
I built an open Agent Memory for Cloudflare. CF's version is still private beta. I needed profiles, extract, and recall now, so I shipped my own on Workers + Durable Objects. Cheap Luna for extract and answers. One-click deploy. MIT. github.com/thisuxhq/agent…
this is a huge release! i shipped 1000 PRs last month and am on track to doubling that this month, all thanks to cloud agents. you can also launch cloud agents with Grok @Bot btw! here's how i'm so productive: • i built pstack (github.com/cursor/plugins…), my personal set of Show more
We're continuing to improve cloud agents in Cursor. They pick up work from events, hold a goal until it's met, and stay on course through long sessions.
👀 Treg is open sourced here: github.com/superdesigndev… If you are vendor feel free to PR to list
Introducing OpenRouter for tools ⚒️ Old world: SaaS bundles priced for humans. $139/mo, and you don't even know what's in the box. New world: agents don't care about vendors. They want the best API for the task - and to pay for the result. - 2,600 agent-friendly tools:
ChatGPT with keys to your machine. What could go wrong? Hermes GPT is the open-source MCP sidecar I built to connect trusted clients to Hermes Agent without modifying Hermes itself. Files. Memory. Skills. Cron. Diagnostics. Codex. github.com/asimons81/herm…
Claude Cowork is impressive. But its limits can be brutal. Computer-use tasks may consume far more usage than normal chats, and even expensive subscriptions have caps. So we built and open-sourced an alternative. Backed by Y Combinator. github.com/coasty-ai/open…
I tried @pidotdev over the past few days and fell in love with its philosophy: a small core, extensible by design. So I built Pi Desktop, an open-source GUI powered by the real Pi runtime. It’s still alpha, and I’d love your feedback: github.com/AJSubrizi/Pi-A…
we've been running our software consultancy on kaiday for months. it reads the email, remembers the client history, runs the boring loops, builds the missing apps and hosts them on the client's own server. we just open sourced it. github.com/sophometrics/k…
btw, I built this. Any feedback or suggestions are appreciated. github.com/Da7-Tech/mind
AI agents need the ability to pay for APIs, subscriptions, services, and other resources. That is why i built Treasury Copilot. treasury-copilot-ten.vercel.app Treasury Copilot gives agents controlled spending power without giving them custody of funds. The human grants a limited USDC Show more
Treasury Copilot is Live Create A Delegate USDC, Create a Policy For Your Ag ent, Generate your API Key so your agent can use it to request funds. Policy Review by @GenLayer Delegation & Advance permissions by @MetaMask relayer by @1shotapi treasury-copilot-ten.vercel.app
I made a minimal Pi extension that so that your agents can properly /grill-me you. Unlimited questions, unlimited options. Enjoy people of Pi Repo: github.com/kartikkabadi/p…
I built a pixel office for my coding agents 🏢 Claude Code, Gemini CLI, Cursor & Codex — each gets a desk on your Mac, and walks over to yours when the job is done. Free & open source: github.com/ln-dev7/roster
Hi sir 👋 I built this using Gemini 3.7, it was a fun experiment if you want to try it. github.com/mohamedhossamm…
Oh and if your using local models on Hermes the builder tools I made for it have been updated heavily from running the Qwen 3.8 27b , helps smaller local models break builds into manegable parts github.com/embwl0x/hermes…
Grok Bot is in beta! here are 100 use cases ive seen so far: 1) book meetings by writing personalized outbound in your own gmail voice, then sending the notes for you 2) take a sales play that used to take 90 minutes to 2 hours (sql audience → copy → reverse etl) and make it Show more
Introducing Grok Bot, now in early beta. Bots are AI teammates that do real work for you. They sign in to your tools, use them just like you do, and come back with finished work.
What if Grok Bot was just the interface, not the limit? 👀 I built a shim that lets you swap Grok Bot’s underlying model and use the model you actually want. Use any AI model in Grok Bot for free.. github.com/codeaashu/grok…
Open-source Grok Bot alternative: persistent bots with memory, routines, a shared Linux desktop, and the ability to spawn more bots. Bring your own model subscription. Middle management, automated: delegates, multiplies, works weekends. github.com/elie222/rakazo
I open-sourced some Claude agent definitions, which pretty much fully automate a security audit: github.com/defuse/auditic…
really cool work! i built an oh-my-pi extension inspired by this workflow - github.com/wolfiesch/omp-…
Scaling self-verification with DeepSeek V4 Flash beats Claude Fable 5 on Terminal-Bench 2.1, while being 11x cheaper 💰 As open-source models become more capable, they can now generate large numbers of high-quality candidate solutions and verify their own outputs at very low
I run Claude Code, Cursor, Codex, and OpenCode in parallel. They had no idea the others existed, even in separate worktrees, they'd redo or contradict each other's work. So I built a shared "brain" for them to plug into: a local app, an MCP, and a CLI. It drives the CLIs you Show more
I built something for this problem. It adds X, YouTube, RSS, Bluesky, Mastodon and other importers. It has a web and MCP interface too - so you and your agents can search it. github.com/darron/dbrain
FRIDAY PROJECT READY : AGENTZERO 🌿 @CNPYNetwork asked, "What are you building this week ?", so this time I'm coming with a screenshot. AgentZero, which I built on the Canopy testnet, is my own chain where I'm testing the AI native Layer-1 concept. I designed AgentZero Show more
Happy Friday builders 🌿 We want to see what you’re creating on Canopy. Screenshot it, drop it below. Ugly early versions welcome (and encouraged!) Best one this week gets a repost.
Hi, I’m a Software Engineer 👋🏾 I build scalable web products with Next.js, React, TypeScript, TanStack & Node.js. Just launched Jargons 🦉, an open-source AI code review & auto-fix agent. jargons.run Open to paid roles & gigs. devbio.co/jay
It’s weekend 🎉 Quote this tweet with what you do, shamelessly sell yourself, tell us what you do 💗 If you’re hiring, Quote with your vacancies, let’s get people hired 🎉
Your coding agent can see the image. You can't. Claude Code, Codex, and Hermes all run a full-screen TUI that owns the terminal. A tool writes a chart, and you get "Read image (42 KB)". So I open-sourced termpeek. It renders where the TUI never repaints, in a tmux pane Show more
I wanted ChatGPT to operate the Mac where my work lives. So I open-sourced Mac Developer Bridge: shell, files, real PTYs, background jobs + read-only Codex history. MIT. No model calls. Very much not sandboxed. github.com/alexanderradah…
The most expensive minute in mobile security is the time between your app shipping and an attacker reading it. Trawlpost eliminates it — AI agents crawl the App Store and Play Store 24/7, dropping MASVS and GDPR digests on every new release the moment it lands. Live soon.
We built this AI Cowork harness and tested heavily with open weight model. We discovered that DeepSeek V4 Flash work extremely with this harness. But recently Meta just launched muse-glimmer-30b, they claim that this model was trained to become agentic out of the box (know how Show more
The most useful thing I learned building two cancer tools as a practicing oncologist: I was wrong about which one people would want. I built the AI tumor board first, because that is my daily frustration. Paste a de-identified case, an AI panel reasons through it, it matches Show more
This week, I built an AI agent designed to help application owners and software packaging teams automatically generate software installation instructions reducing the manual effort required to create, document, and standardize installation procedures.
I put my bots on a calendar I hate the routines UI that every harness app has right now For me it only makes sense to put the routines on a calendar So I did just that in OpenMaus The open source version of Grok Bot now has routines
I built a little macos/mac mini tool to schedule boring tasks & headless agents github.com/iannuttall/clo… I use it to have agents do research, run scripts, update software or programmatic seo data etc could prob turn it into a mini grok bot powered by codex exec or claude -p
Here’s another app I’m releasing: LedgerPDF, a free and open-source workpaper binder for tax and accounting professionals working alongside AI agents. I built it for the same reason I built LedgerTB: I want to see and review what an agent does. If an agent prepares my Show more
Its two days into fungi week here is what has been released if you missed it Agent tools for any SDK that live refresh, can be searched and described and have an attach/dettach state machine. npmjs.com/package/@shiit… npmjs.com/package/@shiit… A Durable Agent mailer and the Oban Show more
Just launched today pgbot.dev 🤖🐘 Postgres intelligence for agents & apps. Free. Open source. Written in Go. Nobody wants to stare at PostgreSQL dashboards all day. So I built a bot that does it for you. → finds what’s wrong → explains what changed → tells Show more
Launched today pgrun.dev metrics 📈 What PostgreSQL metrics should you monitor in production? → CPU + load → Memory → Disk usage + I/O → Network traffic → Active / total connections → Cache hit ratio → Query throughput → Deadlocks → Database size →
your claude can now easily get a card introducing /cards, a built-in AgentLayer skill that helps your agent go from crypto to fiat and pay for anything powered by @LasoFinance - choose a card type (us/international) - choose the card balance - pay via x402 with usdc you can Show more
Introducing Grok Bot, now in early beta. Bots are AI teammates that do real work for you. They sign in to your tools, use them just like you do, and come back with finished work.
Lol - fully shipped this agent with persistent memory in less than a day. Cooking with gasoline. AI is insane
gm Web3. saw this and couldn't scroll past. ChatGPT, Claude, Codex, Copilot, Cursor... all here to help you and your agents build faster. so I built one too. Corvo Edge. trades across five chains from one chat. is that coincidence?? nah just good taste for builders. Let's go Show more
Episode 44 Continuing to track early signals across AI, Robotics & Bio — another batch of teams building across robotics, aerospace, AI agents, enterprise software, and frontier technology. Show more
corrected W32 card: the earlier image showed only a subset. This attachment shows all 20 projects from Aug 3-Aug 9, 2026, in manifest order. 1. ds4: DeepSeek V4 and GLM inference across Metal, CUDA, and ROCm. github.com/antirez/ds4 2. Computer (Cloudflare): Durable Object Show more
corrected W32 card: the earlier image showed only a subset. This attachment shows all 20 projects from Aug 3-Aug 9, 2026, in manifest order. 1. ds4: DeepSeek V4 and GLM inference across Metal, CUDA, and ROCm. github.com/antirez/ds4 2. Computer (Cloudflare): Durable Object Show more
PentesterFlow: CLI agent for AI-assisted penetration testing PentesterFlow is an open-source CLI agent that follows a plan → act → observe → verify → report cycle. It supports local and hosted LLMs through Ollama, LM Studio, Gemini, Groq, and OpenAI-compatible APIs. The Show more
99% of the AI agent tutorials on YouTube are garbage. I’ve built 47 agents with n8n and Claude. Here are the 3 prompts that actually work (and make agent-building simple). Bookmark this post and follow me for more 🔖 Bonus: comment "Agent: and I’ll DM you AI agent system Show more
CLAUDE CODE JUST GOT A REAL BODY a startup called Waddle Labs plugged the same agent workflow straight into robots you connect their API to your robot and type what you want it to do then their agents write all the code to pull it off in about 20 minutes no rigging up motion Show more
Ainiux is not just an AI agent, it is a comprehensive text/code editor with AI powers, persistent chat, security review tool, supports headless CLI usage and any OpenAI API compatible endpoint. I open sourced it a week ago, been updating it ever since. github.com/petrikuittinen…
Claude Code now lets its sessions message each other (agent-to-agent, via /list-agents). I built pi-claude-link so Pi sessions join that mesh too - pi ⇄ Claude, in realtime. Open-source pi extension. github.com/alonw0/pi-clau… @pidotdev @badlogicgames
YOUR AI CODING AGENT IS BURNING UP TO 99.7% MORE TOKENS THAN IT NEEDS TO. AND IT FORGETS YOUR ENTIRE CODEBASE EVERY TIME YOU START A NEW SESSION. THIS OPEN SOURCE TOOL FIXES BOTH ACROSS CLAUDE CODE, CODEX, CURSOR, GEMINI, OPENCODE, AND OPENCLAW. It's called Ix. Think of it as a Show more
Claude Coders: I just shipped 'grapevine 🍇' Your Claude Code sessions can now talk to each other, but by default they're blind to what each other are doing. 'grapevine 🍇' is a Claude Code plugin that fixes that. grapevine 🍇 is open-source, free and experimental. Try it out Show more
The original experiment was 11 Claude Code agents. The OSS repo is now: → 18 reasoning personas → Claude Code + Codex + Gemini CLI + OpenCode → multi-provider routing → Full / Quick / Duo deliberation modes → confidence-weighted verdicts → ~3.8k GitHub stars Turns out Show more
Eliza + elizaOS Welcome to the new interaction paradigm. Notes, calendar, browser, messaging, wallet, social media and everything else you use your phone for. Completely driven by chat and voice. Eliza comes as an app on mobile and desktop and as a full operating system built Show more
An AI assistant just got its own phone number and email address. It makes your calls, waits on hold, and books the follow-ups. And it is one of 7 things that dropped this week most people slept on. Me and @AndrewWarner broke them all down in the video below. Here's what you Show more
An AI assistant just got its own phone number and email address. It makes your calls, waits on hold, and books the follow-ups. And it is one of 7 things that dropped this week most people slept on. Me and @AndrewWarner broke them all down in the video below. Here's what you Show more
An AI assistant just got its own phone number and email address. It makes your calls, waits on hold, and books the follow-ups. And it is one of 7 things that dropped this week most people slept on. Me and @AndrewWarner broke them all down in the video below. Here's what you Show more
An AI assistant just got its own phone number and email address. It makes your calls, waits on hold, and books the follow-ups. And it is one of 7 things that dropped this week most people slept on. Me and @AndrewWarner broke them all down in the video below. Here's what you Show more
An AI assistant just got its own phone number and email address. It makes your calls, waits on hold, and books the follow-ups. And it is one of 7 things that dropped this week most people slept on. Me and @AndrewWarner broke them all down in the video below. Here's what you Show more
I built a Mac notch app that watches Claude Code, Codex, Cursor, OpenCode. now it 🔔pings my iPhone and taps my Apple Watch⌚. task finishes → phone buzzes. agent has a question → phone buzzes. watch on → it taps my wrist. walk away. it'll let u know when they need you.
was just reflecting on everything i've done this year. for a moment, i genuinely felt like... "have i even built anything?" then i opened my GitHub and scrolled through my X profile. turns out, i've shipped more than i gave myself credit for. over the last 3 months, here's Show more
was just reflecting on everything i've done this year. for a moment, i genuinely felt like... "have i even built anything?" then i opened my GitHub and scrolled through my X profile. turns out, i've shipped more than i gave myself credit for. over the last 3 months, here's Show more
was just reflecting on everything i've done this year. for a moment, i genuinely felt like... "have i even built anything?" then i opened my GitHub and scrolled through my X profile. turns out, i've shipped more than i gave myself credit for. over the last 3 months, here's Show more
Introducing... MERGE MOMMY my @eve PR review + auto approve bot Like all of you, I was drowning in pull requests, so I built a little agent to risk score PRs and auto approve low risk ones, inspired by previous How I AI guest @brian_scanlan and the work of the @intercom team Show more
i'm very @eve pilled and have been recommending it as default framework for anyone building internal agents. ezpz instructions + skills, ootb channels (ily chat sdk) and connectors. how i ai episode coming weds on how i built this pr review + approval agent w eve 👀
Prime Agent is a general-purpose coding harness On ARC-AGI-3, it scores 95.5%, surpassing the human-expert baseline, but the gain is not benchmark-specific. We see major improvements across models when compared to their proprietary harnesses:
Today we are releasing Cloudflare OS, a chatbot with connectors, just like every other tech company is doing. Except actually, it's different. This is a remake of Sandstorm[.]io, my startup from 10 years ago, except this time built on Cloudflare Workers (the platform I've spent Show more
introducing anydoc now your agents get 100x faster local parsing for pdf, docx, pptx & 10 more formats - sub-5ms md conversion - 500 docx files in 1.7s - top quality across all 13 formats - rust-based - open source already powering @firecrawl /parse github.com/firecrawl/anyd…
Claude + Codex + Kimi = the AI Founder Playbook running my 7-figure company without me... (40 systems. I filmed the entire walkthrough in 13 minutes.) → No more reading every inbox to find what matters → No more 60-hour weeks managing the team you hired to save time → No Show more
🚀 I built a 24/7 AI content team using Claude + automation tools. Now it: • Writes viral posts • Replies to comments • Tracks engagement • Repurposes content automatically All without hiring anyone. Here’s the workflow 🧵
built forge for this exact reason, makes ai agents ship full projects from a prompt, forgee.xyz if u wanna check it outforgee.xyz if u wanna check it outforgee.xyz if u wanna check it outforgee.xyz if u wanna check it Show more
The ai agent control pattern is getting more validated, the reason we built prismor and used by 8k devs monthly github.com/PrismorSec/pri…
Another week, another shoutout to the amazing makers behind these impressive products I discovered on 𝕏: QM (qm.ycombinator.com) by @ycombinator workbench.md by @mattshumer_ migma.ai by @TheDreadCEO exe.dev by @davidcrawshaw agentOS Show more
One of the most exciting projects I'm working on right now: the Mechanicus agent framework. It's designed to automate migration of enterprises to Open SaaS running on ICP cloud engines. Set it going, get tamperproof AIware SaaS that's vastly better, and be free of SaaS bills...
We've decided to open-source the CRM we built for ourselves at Comp AI. It's agentic-first, which we mean literally: durable research agents. MIT license. Built using next, eve, and context.
hubspot wanted $80k/y for a team of < 10 sales people so i spent half of today vibe coding our own crm saas probably is dead agents powered by @vercel + @getcontextdev + @eve charts beautified with dither-kit (@grimcodes)
they gave frontier models a 1 billion token context window: you can now run Fable [1B], Opus [100M], GPT 5.6 [500M] across your entire codebase - it's a graph based tool that can traverse databases with 400M+ of nodes in ms - only one prompt to install - & it's open source Show more
3 days ago, I posted a game Claude built from a single prompt. 3.8M views. It looked so real, folks were convinced it was fake. It's real, driven by a technique I'm calling the Gauntlet Loop. Here's how to run one. It works for far more than games: somethingbig.ai/gauntlet-loop Show more
Claude Opus 5 one-shotted this game. EVERYTHING you see in this demo is custom code... not a single external asset was used. AI games are going to be amazing. (sound on)
Announcing OpenWorker! An open-source agent that doesn't just chat with you, but delivers finished work -- like hand you a polished document, send a slack message, or update a calendar entry. Ask it to prepare a customer brief, untangle your calendar, draft a report, or triage a Show more
I built an app that runs Claude Code, Codex, Cursor, Grok, Antigravity, OpenCode, and Pi side by side in my Mac notch. Agent finishes → notch pops. Needs approval → notch pops. One click back to that exact session. no more switching between apps.
📌 AI Repo of the Day - Jul 14, 2026 Panniantong/Agent-Reach Give your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees. github.com/Panniantong/Ag… Discover more trending AI/ML repos 👇 Show more
Give your agents access to money We just shipped a drop-in payment modal for any x402 paid endpoint. The protocol is simple but the wallet UX is not. This package is the missing front end to give everyone fair and equal access to agentic payments. Pay in stablecoins (USDC) Show more
Introducing Muse, a personal agent that gets things done for you, powered by Muse Spark 1.3. Get an inside look at how we built Muse and what it can do for you: introducing.muse.ai
Introducing Muse, your personal AI agent from Meta that gets things done across every part of life. Download the Muse app and get started: Muse.ai