Episode 117: AutoDev Git Diffs We extend our HUD UI to show beautiful git diffs of code changes made by AutoDev.

Episode 107: Codebase Indexing via Greptile We write an @extism plugin for @greptileai and consume it via our new Golang/HTMX frontend for AutoDev. Claude Artifacts makes it fast! - Greptile Extism plugin: - AutoDev UI:

Episode 108: GitHub File Explorer We add a GitHub file explorer to our AutoDev agent workspace. We review AutoDev's agentic loop and the building blocks we'll make next.

Episode 109: Exploring OpenDevin & CodeAct We learn how OpenDevin implements the CodeAct paper for agent planning and task execution. Background 📜👉

Episode 110: AutoDev Planner We ask AutoDev to make a plan and it does.

Episode 111: Heads-Up Display We give AutoDev a heads-up display (HUD). We use voice (via @GroqInc) to search over our codebase with natural language (via @greptileai), seeing a summary of each file in a separate HUD pane.

Episode 112: Using AutoDev We review our new AutoDev product and use it to build a 3D canvas for a knowledge graph of codebase insights.

Episode 113: Agent Memories & Reflections We use AutoDev to implement the memory, retrieval and reflection systems from the Generative Agents paper. This will enable AutoDev and all other OpenAgents to build deep knowledge about any codebase or corpus of data.

Episode 114: Planning & Execution We upgrade AutoDev's planning to include additional context from past memories and relevant files. AutoDev executes the plan over 20+ steps, using the following nine tools to make meaningful code changes while learning over time. Next up, submitting pull requests to GitHub! - view_hierarchy - View file/folder hierarchy at path - view_file - View file contents at path - find_files - Find relevant files to query - remember_insights - Perform a similarity search on previously recorded insights - query_files_by_vector - Query files by vector similarity based on their summaries - generate_plan - Generate a multi-step plan for complex tasks. - retrieve_memories - Retrieve relevant memories from the memory stream - add_memory - Add a new memory to the memory stream - reflect_on_codebase - Generate higher-level insights about the codebase

Episode 115: AutoDev as Pair Programmer We use AutoDev to edit the AutoDev codebase via the GitHub API. We see the changes in seconds. The workflow is way faster than copy-pasting from Claude Artifacts!

Episode 116: 3D Multiplayer We demo an alpha build of the world's first Agent MMO.

Episode 118: Version Two We say goodbye to ChatGPT, Claude, Copilot & Cursor! We introduce version 2 of OpenAgents: your all-in-one productivity dashboard. v2 includes: ➡️ Chat with top AI models ➡️ Multiple simultaneous chats in draggable/resizable windows ➡️ Deep integration with any GitHub codebase: create/edit files, smart search via @greptileai, open pull requests ➡️ Web scraping via @firecrawl_dev ➡️ Pay-as-you-go via Stripe or Bitcoin ⚡️ 👆That all goes live to the public on Monday, August 12. And coming soon: ➡️ File uploads ➡️ Voice commands ➡️ Long-term agent memory ➡️ Long-running agents ➡️ Third-party plugins ➡️ Agent Store ➡️ Rev-share referrals ➡️ Making all Microsoft products irrelevant starting with Copilot, VSCode & GitHub Stay tuned!

Episode 119: v2 Beta Launch We demo our new v2 chat & auto-coding interface, live now at All users can now use all features for free with GPT-4o mini, or buy pay-as-you-go credits to use advanced models like Claude 3.5 Sonnet. For a limited time, anyone can DM us here on X to receive $10 of credits for premium models. (Send us your username or email you signed up with.) We are eager for feedback and feature requests! Please try it out and tell us how it can be more useful for you. We have an ambitious roadmap with much more to come. And thanks to v2 here, we build faster than ever. 😎 Stay tuned!

Episode 120: Exploring SWE-bench Verified We talk smack about benchmarks but conclude they may finally be worth our time. We do a dramatic reading of @OpenAI's blog post then feed it to OpenAgents which sets up a new repo as benchmark workspace. We're going for the high score!

Episode 121: SWE-bench Planning We make a plan to win high score on the SWE-bench Verified benchmark. We pull the 500 samples into a web UI for easy inspection -- super smooth thanks to @convex_dev! -- then decide to focus first on the psf/requests repo. Next we index!

Episode 122: Codebase Indexing via GraphRAG We design a codebase indexer using GraphRAG, a new graph-based RAG algorithm from Microsoft. We ask OpenAgents to run a basic version of this on our v2 codebase, identifying 24 topical 'communities' and associated files & functions.

Episode 123: GitHub Issues to Pull Requests We solve multiple GitHub issues without touching a line of code. Our coding agent works on four issues simultaneously, submitting multiple pull requests within seconds of each other. The future of coding is not in your editor.

Episode 124: Magic AI = Deep State AI coding startup @magicailabs raised $465M with the stated goal: "Imagine if you could spend $100 and 10 minutes on an issue and reliably get a great pull request for an entire feature. That’s our goal." But we shipped that two days ago, our agent solving two PRs for in 12 minutes for $5 in a product that's publicly available now everywhere in the world: ( So what does Magic really need all that money for? Reviewing their latest statement calling for AI regulation and their investor list, including deep state financier @ericschmidt and @natfriedman /@danielgross, primary funders of "nationalize the AI labs" @leopoldasch, it's clear they want to build AGI and hand it to the deep state. Magic, SSI, OpenAI and Anthropic have all clearly aligned with ClosedAI to the benefit of themselves and governments-- to the detriment of all humanity. We will raise a coalition to counter their influence and smash every nexus of centralized corruption. The world must have truly open AI. How will you help?

Episode 125: The Master Plan We detail our master plan: 1. Build the best agents 2. Sell agents and pay you a cut 3. Repeat 1 and 2 until world conquest We're building a universal graph for agent knowledge and execution paths in which contributors earn bitcoin revenue-sharing proportional to usage. ⚡️ We're bootstrapping this network by first building custom agents for small businesses. Soon we'll expand into a self-serve platform in which agent builders & buyers can connect in an open marketplace. OpenAgents will have the best agents because we will pay agent builders the most.

Episode 126: OpenPress and the End of Wordpress Our agents are hungry for work! We notice that 43% of all websites are powered by WordPress, a supposedly open-source & community-driven framework in fact controlled by an extortionate megalomaniac thug. Maybe some room for 🤖 improvement there! We review the WordPress drama, brainstorm a lightweight alternative from scratch called OpenPress, set up a fresh GitHub repo, and install @laravelphp. Stay tuned!

Episode 127: Hello OpenPress We build a basic @laravelphp blog and launch the first OpenPress site to First post 👉 Code 👉

Episode 128: Styling OpenPress We restyle OpenPress using @shadcn UI components, using OpenAgents to convert them from React to @laravelphp Blade. Starts to feel like a real blog!

Episode 129: Hosting OpenPress We spec out our initial plan for OpenPress hosting: 1. A managed service at similar to 2. Self-hosting available always from our open-source (CC-0) repo at We're coming after that @automattic AND @wpengine money 😂

OpenPress now has its own video series which continues here:

Episode 130: OpenAgents <> OpenPress We summarize the WordPress situation and how @OpenPressAI fits into our master plan. We dust off our open-source GitHub repo from v1 and create a new Laravel app for v3, which we'll rebuild in ~1 week using code to be shared with OpenPress.

Episode 131: v3 Landing Page We build a simple 'coming soon' landing page from components converted from shadcn-ui to Laravel Blade. We go cross-eyed at the vanilla CSS so break down and add Tailwind, but otherwise stick to basic Blade & no-build JavaScript. Can we launch v3 without a package.json?! Code 👉

Episode 132: v3 Core Feature We design and start building our core feature largely from scratch: "Ask for something, an agent makes it happen." We implement our main data models (User/Team/Project/Thread/Message), build our chat input form, and save messages to the database.

Episode 133: HTMX Server Sent Events, Part 2 We add streaming to our chat UI via HTMX server-sent events. Drawn back into the inescapable orbit of @htmx_org, we make fun of Vercelbros who can't stream multiple simultaneous messages in a single chat like we can with HTMX.

Episode 134: Inescapable Inertia We review our HTMX/vanilla integration attempt, conclude Laravel+React is the right stack for OpenAgents v3 & @OpenPressAI, and install Inertia v2 in a fresh Laravel app in our main open-source repo. Code 👉

Episode 135: Agentic CRM Design We need a CRM! And why use ancient bloated enterprise slop like @salesforce when we can whip up a quick one ourselves, with perfect integration with the rest of OpenAgents? Agents mock up a UI, write a spec, and implement our first 10 unit tests.

Episode 136: Teams Speaking of ancient bloated enterprise slop, let's replace @MicrosoftTeams too! We now have multiple small businesses wanting to share chats/files/agents between multiple users. Agents build our team creation UI and connect our team switcher menu.

Episode 137: Projects Next we emulate Claude Projects. Their UI is great but why limit yourself to models from only one AI lab? We want to use any model or agent with our projects. Agents build our initial project create+view UI and relevant tests.

Episode 138: Year One Recap We recap the first year of OpenAgents and look ahead to year two. Topics: - What went right & wrong with our agent store launch - Our pivot to business customers and a tour of our new 'pro' product - The upcoming relaunch of our agent store and bitcoin incentives A few highlights: 👇 --- OUR THESIS: OPEN AGENTS FOR EVERYONE "Our core thesis is intact. I expressed in the opening video of this series, we now agree with Sam Altman that the future would be agents. Agents are the thing that businesses and people will be paying for. And to take that one step further, we believe that every person, every business, every family, every street corner, every device is going to have one or more AI agents working on your behalf or as part of the system or device or physical location. Agents will be everywhere. Do you want these massive closed-source corporations with poor incentives, incentives to get their tendrils around the government, scaremonger about AI safety and try to erect barriers to keep out startups? We cannot let that happen. The stakes are too high to mess with that level of interference with what should be an organic, bottom-up, evolutionary process." --- BITCOIN/NOSTR & INCENTIVIZED INTEROPERABILITY "Incentivized interop matters. What we had with 20 different developers coming to write plugins [and earning bitcoin revenue share], I could see very clearly that this is the future of how we build this unified agent substrate that's usable by all people. Obviously this idea of an operating system for all businesses is probably larger of a scope than any single person or company is going to be able to take on, but we are certainly positioned at OpenAgents to provide a leading role in that in part because we have an answer to this question of incentivized interoperability." "How you incentivize participation is you pay people. You pay people what? Money. What do you not pay them? You don't pay them a fucking shitcoin chuck-e-cheese coupon that somebody just made up. You pay them real money. How do you get real money? Customers pay you for it. Not that everyone needs to pay us in bitcoin, you can pay us in dollars and we'll pay out in bitcoin." --- We'll be opening up our v3 pro product in the coming weeks to more business users, preparing for a broader public launch along with the relaunch of our agent store in the coming months. DM us for early access or fill out the inquiry form on openagents dot com. A massive thank you to all our users and supporters and everyone who's given us feedback or retweets or attention or anything else. ❤️ Much more to come! Stay tuned 🤖👍

Episode 139: Going Mobile We're launching a mobile app in two weeks! We'll make the launch fun by copying the "12 Days of ClosedAI" 😂 with our own "12 Nights of OpenAgents" 🎉 We'll release one video every weekday night from now through Dec 20 exploring Onyx features, architecture, Bitcoin/Nostr integration & more. By night 12 we will launch "the coolest, awesomest, bestest, open-source, truly open AI agent app and product that should bootstrap a global decentralized AI agents marketplace running on open protocols, accessible to everybody -- and you don't even have to pay us $200/month." Night 1 begins now with a sneak peek of the app - called Onyx - and an overview of the what, why & how.

Episode 140: Open-Sourcing Onyx We open-source Onyx and tour the docs: Because the agent economy must be built on a foundation of openness, not black-box bullshit with unverifiable claims and code!

Episode 141: One Market We share the big-picture vision of OpenAgents: to liberate frontier AI from captured centralized corporations by building one decentralized global marketplace of AI agents and services. Onyx will be your gateway to that marketplace. It's an easy-to-use mobile app that connects your voice commands to a broad ecosystem of service providers competing to execute your commands, all powered by open protocols unstoppable by centralized gatekeepers. We will not allow the future of AI to be controlled by the likes of Microsoft+ClosedAI and their government handlers. The agent economy must be built on a foundation of openness. Here's our plan to do it.

Episode 142: Data Vending Machines We demo Nostr's "data vending machines" protocol (NIP-90) in Onyx, requesting LLM inference that's then handled by a separate service which processes DVM inference jobs with Llama 3 via the @GroqInc API. The big idea: We can make any AI services available through an open, neutral, decentralized network, all accessible through one easy-to-use mobile app. Up next, we'll combine this with seamless instant micropayments so service providers can earn a steady stream of bitcoin! Further reading: - - - Demo backend repo: -

Episode 143: Onyx as Bitcoin Wallet We demo sending & receiving bitcoin in Onyx. Onyx now has a simple self-custodial Bitcoin/Lightning wallet thanks to the @Breez_Tech SDK. This will make it easy for anyone in the world to earn and spend bitcoin for AI agents & services. Further reading: - (We're using it!) - (We're using it!) - (We're testing it!) - (The code)

Episode 144: Pylon and the Model Context Protocol We introduce Pylon, the node software that powers Onyx. It's a desktop app anyone will be able to run, and will launch alongside Onyx later this month. We explore the new Model Context Protocol (MCP) from @AnthropicAI, an open protocol that helps build agents and complex workflows on top of LLMs. Pylon will serve as both MCP server and Nostr DVM service provider, enabling node runners to earn bitcoin from our AI marketplace while also using it as a backend for their own Onyx app. We demo Onyx -- now the first MCP mobile client -- connecting to Pylon, now the first MCP server written in Rust via @TauriApps. Further reading: - MCP announcement: - MCP site & spec: - Pylon repo: - Tauri:

Episode 145: Going Local Onyx now runs Llama 3.2 3B locally on your phone - enabling free and private chats! You can also chat with the larger Llama 3.3 70B by connecting via MCP to a local Pylon node, which can load any model supported by @ollama. Check the video for demos of both. First alpha builds launch in ~1 week for Android and iOS. With a mix of on-device, local or distributed servers, and cloud APIs, we have a LOT of flexibility for building agentic workflows using any combination of tradeoffs for privacy, speed, cost, and censorship resistance. Combined with our DVM marketplace from Ep. 142 and bitcoin wallet from ep. 143, we've got everything we need to bootstrap a decentralized marketplace of agentic AI services. That wraps up our first full week of '12 Nights of OpenAgents'. See you Monday!

Episode 146: Sensemaking: Drones What's up with those drones? We ask ChatGPT, Perplexity & Grok -- and get no good answers. We try to improve on their answers in a single day of building our own search using basic versions of GraphRAG and test-time compute, enabling the agent to 'think' over a knowledge graph before generating a response. We get that full flow working with preliminary data, generating a decent response - but not yet better than the billion-dollar AI labs. Maybe in another day or two? We need to feed it better data! We'll do that tomorrow.

What’s the cost of mistakes in your contracts? If you work with contracts day-to-day, it’s time to automate. Track every detail, streamline workflows ... ✨ Make managing contracts as easy as a few clicks. Visit our new website & book your demo today!