screenpipe (YC S26)

introducing screenpipe: it records and learns how you work and turns it into a searchable memory, SOPs, and AI agents open source, local-first, 20K+ GitHub stars, 1,900+ forks, and 130+ contributors

meeting note + screen recording = 🪄 @meetgranola, @PLAUDAI @otter_ai @NotionHQ @bee__computer we're also integrated with all of those so you can have your meeting summarized from your screen and their transcriptions

where did your day actually go? screenpipe tracks how you really spend your time -every app, every task, on your device - and writes the breakdown + your daily journal automatically.

In professional services, your margin is how fast your people deliver. But on every engagement, your best consultants re-solve the same problems from scratch, and that hard-won know-how walks out the door when they leave. So we built screenpipe for firms 👇

most AI assistants are brilliant interns with amnesia ask about your own work → they've got nothing screenpipe gives your AI a memory of everything you've seen, said, or heard - on your device - so it answers from what actually happened

I just joined Y Combinator Summer 26. I built 200+ open source projects before (without AI), here's my story:

500 tabs and 50 apps open, 3 laptops, don't know what you did at the end of the day? screenpipe records your screen + audio locally and lets you search your whole week in plain english - then jump straight to the exact moment it happened 100% on-device. open source

@samdenty rust: github.com/screenpipe/scr…

There’s an OSS implementation of the codex permission flow in Swift here: github.com/zats/permiso

GitHub 爆火项目榜：本周新增星标精选 TOP 7。 我最喜欢推荐7 screenpipe 最符合我对 AI 时代个人工具的期待，它不是一次性玩具，而是能持续积累价值的“第二大脑”。 这些项目在本周星标增长明显，涵盖 AI 视频生成、虚拟伴侣、安防监控、物理仿真、照片管理等领域。 1 MoneyPrinterTurbo +7948 【造富潜力】 仓库：github.com/harry0703/Mone… 简介：利用 AI 大模型，一键生成高清短视频。输入主题或关键词即可自动完成文案、素材、字幕、配乐并合成完整视频，支持中英文。 亮点：全自动流水线 + WebUI，适合内容创作者快速生产。 造富潜力：项目名称本身带有“印钞机”意味，社区常将其视为批量生成短视频的工具，适合尝试制作无脸视频或带货内容并上传抖音、YouTube、TikTok 等平台进行流量变现。但实际效果高度依赖视频质量、平台算法、内容合规以及个人运营能力。目前社区讨论多为潜力宣传，尚未有大规模公开验证的成功收益案例。 2 airi +1079 仓库：github.com/moeru-ai/airi 简介：自托管的 AI 虚拟伴侣，支持实时语音聊天、可玩游戏（Minecraft / Factorio 等），兼容 Web / macOS / Windows。 亮点：完全本地部署，可自行控制。 3 Frigate +1308 仓库：github.com/blakeblackshea… 简介：实时对象检测的本地 NVR，专为 IP 摄像头设计，集成 Home Assistant，使用本地 AI 完成物体识别。 亮点：隐私优先，无云端依赖，智能家居用户常用。 4 genesis-world +247 仓库：github.com/Genesis-Embodi… 简介：通用机器人与具身 AI 的仿真平台，集成多物理引擎和照片级渲染器，用于物理 AI 开发。 亮点：生成式世界模拟，适合机器人和强化学习研究。 5 ppf-contact-solver +1603 仓库：github.com/st-tech/ppf-co… 简介：物理仿真接触求解器，支持布料、实体和杆的复杂物理模拟。 亮点：高性能接触求解，适用于游戏开发、材料仿真等领域。 6 PhotoPrism +68 仓库：github.com/photoprism/pho… 简介：AI 驱动的自托管照片管理应用，类似私有版 Google Photos，支持自动标签、智能搜索和 RAW 处理。 亮点：完全本地运行，隐私优先。 7 screenpipe +146 仓库：github.com/screenpipe/scr… 简介：本地屏幕 + 音频持续录制工具，24/7 捕获数字生活，生成可搜索的 AI 记忆库。 亮点：把电脑变成有记忆的个人 AI 引擎，支持后续个性化工作流。

My AI employees: - ChatGPT - Project manager - Replit - Developer - Grok - Market research - Manus - Secretary - OpenClaw - COO - Higgsfield - Video producer - Gemini - Graphic Designer and Editor (of images and files in GDrive) - Origami - SDR - Wispr - Stenographer - Screenpipe - All-knowing sage - G.E.N.A. - can’t talk about this yet Total monthly cost ~$720 (before paying for additional credits as needed) This isn’t the entire stack I use, but it’s the main ones. Others I didn’t list—Midjourney, Apollo, Gamma, Siri, Plaud, Canva, Alexa, Claude, Claude Desktop, Codex, Cursor, Google (in search—I use this a decent amount), some of my own AIs that I’ve created using various APIs that tie into my private custom tools

i just lost a billion dollars 3 years ago Cursor CEO dm'd for a founding engineer role and I ignored him because I wanted to build my own startup hold me accountable x, I'm gonna make more than a billion with screenpipe

The world is about to become hyperconnected and cross the percolation threshold. Thinking about 1:1 AI replacing human work is peak midwittery, classic SF technofaghitree. AI is about restructuring the sociocybernetics of all civilization and human interaction/connectivity.

quick break down of screenpipe github ⭐️

screenpipe users burnt 200B tokens in the past 30 days

we support cloud pipes - agents that work off your org screen recordings if you need higher privacy, we support only sharing part of the data, with custom data controls, and on premise

introducing: screenpipe enterprise everyone on your team records their screen all the data is centralized in one place then hundreds of agents work off it 24/7, each already knowing how your company actually operates don't hire 500 people, record 12 and hire 500 agents

we're hiring a Head of Tokenmaxxing Revenue stand up a fleet of agents, burn tokens on purpose, turn it into revenue you have to be great with humans and great with tokens sf, in person if you had an unlimited token, what is the first thing you would do to make money?

how to use screenpipe, start to finish. record everything on your screen locally, then search it, replay it, and ask ai about anything you've seen, said, or heard. setup, timeline, pipes, meetings. the whole walkthrough in one video.