Launching today

SecondOS

Launching today

One memory for your codebase, every AI tool reads it

2 followers

One memory for your codebase, every AI tool reads it

2 followers

•

•

Your AI re-reads the whole project every session. SecondOS gives it a memory — so it picks up where you left off, reads ~5× less, and your source never leaves your machine. One memory, every tool.

Free Options

Launch tags:Developer Tools•Artificial Intelligence•GitHub

Launch Team / Built With

Battery AutopilotPut your MacBook battery on AI autopilot

Promoted

Maker

📌

Hey Product Hunt 👋 I'm Oğuz, I built SecondOS. ▎ ▎ I kept re-explaining the same repo to Cursor, then Claude Code, then Copilot, every single session. So I built one memory that all of them read. ▎ ▎ You sync a repo once. SecondOS builds a structure-aware map of it (symbols, signatures, routes, a call graph resolved across TypeScript, Go, Python, Rust, Java) plus durable memory for decisions and gotchas. Any AI tool then reads the same map over MCP instead of re-scanning your files. Across five real repos that is roughly 5 to 9 times fewer tokens to answer the same question. ▎ ▎ Two things I care about that "add memory to your agent" tools skip: ▎ ▎ 1. It is living. A note is anchored to a real symbol. When that code changes, the note is flagged stale instead of quietly lying to your next agent. (There's a live demo of exactly this on the site.) ▎ 2. Privacy is stated honestly, not marketed. Code uploads as a map, the function bodies stay on your machine. Prose docs do upload as full text (a real trade-off, or use local mode with no login). And the client is source-available so you can read exactly what leaves it: github.com/Preatan/secondos-client ▎ ▎ Free tier is genuinely free, no card, local mode unlimited. Pro is $10 a workspace. ▎ ▎ Try it: bunx @secondos/cli init · https://secondos.app ▎ ▎ I'll be here all day. I'd love feedback especially on the drift/staleness model. You can also reach me on X @grbuzoguz.

Report

7d ago

Forum Threads

p/secondos

•

2d ago

We had 40 decisions in our codebase. 4 of them were written down.

I ran a mapper over our own repository this week and counted something uncomfortable: how many of our decisions were ever written down anywhere a person could actually find them.

22 modules. 40 identifiable decisions. Of those 40:

4 were in a real document in the repo

3 were explained in a PR body

p/secondos

•

5d ago

Stop re-explaining your codebase to every AI tool

Hey Product Hunt

I built SecondOS because I kept re-explaining the same codebase to Cursor, then Claude Code, then Codex, in the same afternoon. Every new session, every new tool, back to zero. Your AI has no memory of your project, and you've probably just accepted that.

SecondOS fixes that. You sync your repo once. It builds a structure-aware map of your code (symbols, routes, a real call graph across TS, Go, Python, Rust, Java) plus a layer of durable memory: the decisions, the gotchas, the "this looks wrong but it's load-bearing" that never actually lives in the code. Then every AI tool reads the same thing over MCP. Cursor, Claude Code, Codex, ChatGPT. One memory.

Three things that make it different from bolting "memory" onto an agent:

View all