Your agent should not start from zero every time.
Paste everything into chat and the agent loses focus. Workbench keeps a living memory of your project, so the agent stays oriented without you pasting notes into every message.
Today, most AI coding sessions forget the architecture the moment the chat ends.
Session
Where the work happens. You steer, the agent builds, and what it learns is saved back.
Knowledge
Your notes. Open, fix, and keep them forever yours.
Brain
The map. See how the pieces connect before you dig in.
Knowledge
Your notes. The agent can't rewrite them without you.
Open your project notes. Fix a wrong fact, add what only your team knows, and keep company, org, and repo memory in one place you control.
- Correct a wrong fact the moment you see it
- Keep shared memory for the team, not buried in one chat
- Sync notes to GitHub when you choose to
Company
Defaults and playbooks everyone starts from
Team
Shared rules across the repos you own
This repo
What is true for this codebase alone
Notes the map understands
Every note carries a few simple details. Workbench uses them to place the note on the map and to decide which notes the agent sees first.
| On every note | What it gives you |
|---|---|
| What it is | A service, a concept, a repo, or a rule. Workbench knows what kind of thing each note describes. |
| Name | A stable ID behind the scenes, so renaming a note never breaks the links to it. |
| Main topic or detail | Big topics become hubs on the map; details hang off them. |
| Other names | Teach it that “frontend” and “UI” mean the same thing, so both find the right note. |
| How much it matters | Mark the notes that matter most so they reach the agent first. |
| Where it came from | The files that were read to write the note, so you can check it. |
| Links | How notes relate: part of, uses, or related to. |
| Disagreements | When two notes contradict each other, Workbench keeps it visible instead of hiding it. |
Sketch
Start from the repos in your system
Deepen
Let the agent fill in the main topics and how they connect
Explore
Ask about a concern and follow the trail
Brain
See the project. Then ask around a concern.
Brain turns your system into a readable map: services, standards, frontend, DevOps, and the links between them. Build it once, then explore any question without losing the big picture.
- See how frontend, backend, and ops actually connect
- Grow the map before a big feature, not after the mess
- Ask around a topic and follow the path that matters
Two surfaces. One memory.
Knowledge is where you read and write. If the map is wrong, you fix the entry.
Brain is where you look. If you need orientation, you open the map.
Build memory before the rush
Teach the project
- 1.Open Brain and pick the system you care about
- 2.Grow the map until the big pieces show up
- 3.Explore a topic: frontend, deploy, or standards
- 4.Fix anything wrong in Knowledge
Stay in the session
Just ship
- 1.Start work like you always do
- 2.The agent gets only the notes that fit this job
- 3.Finished sessions save what they learned automatically
- 4.Open Knowledge any time to check or correct it
Where your project memory shows up
Knowledge is not a separate tool you have to remember to open. It feeds every part of the work.
01
Every message
Each time the agent works, the few notes that matter for that step go with it.
02
The plan
Plans are written with your company rules and project context already in them.
03
The quality check
The work is checked against the plan and your project's quality notes.
04
Switching agents
Change agent mid-task and a running summary of the session goes with it.
05
Full system review
Reviews use the map to catch changes that would break other repos.
06
After every session
Finished sessions and linked pull requests save their key facts back automatically.
While you work, the right notes show up.
Each step gets a small, focused set of notes, not everything you have. When a session finishes, its key facts are saved back, so the next one starts smarter.