Every agent, fluent in your project.
Stele gives the tools you already use a shared memory of your decisions, tasks, and lessons.
$curl -fsSL https://app.stele-ai.dev/install.sh | bashThree commands, then it's automatic.
No dashboard to configure and no workflow to learn. You set it up once, then keep prompting your agent exactly the way you do today.
Install & sign in
One line installs the stele CLI and signs you in with GitHub. No account to create.
Run /stele:start
Inside your agent, run it once. Stele wires into your project and builds the first record.
Keep working
That's it. Your agent now reads the record before it answers and writes back what it learns.
Copy & run — that's step one
macOS, Linux & Windows · one binary, wired into your shell.
The record catches what a fresh agent can't.
One small moment that happens in every project with more than one contributor. Here's how it goes once the reasoning lives in the record.
A harmless-looking change reintroduces a problem you already paid for.
The linked risk surfaces while the work is happening — not after it ships. The agent pauses and asks instead of guessing.
See more examples in the docs“This lock looks unnecessary — I'm removing it.”
That lock came from TASK-91 during the auth rework. Removing it reintroduces ⚠ KNOW-118, a race condition that reached production once. Want me to keep the guard and simplify around it instead?
Give your project a memory in three commands.
The decisions, the tasks, and the lessons behind your code — in one place your future self and your agents can both read.
Sign in with GitHub · free during beta · or read the public commons first