L2 The agent loop
Stopping
Ends a run in a clear, named way, hands back the work done, and checks that "done" really means done.
Why it matters
An agent that stops silently loses work the user paid for. One that declares victory too early hands over half-finished results as if they were complete.
What to build
One typed outcome with work saved per exit
The default the research recommends. The reasoning and the code behind it are in the app.
What it takes, by kind of agent
- Conversational assistantMust-have
- CodingMust-have
- Deep researchMust-have
- Browser/GUIMust-have
How well it is studied
- Harnesses checked
- 30
- Checked facts behind it
- 376
See the 376 checked facts and how each harness does it
The facts, the code lines they come from and the rules for this topic are in the app. Sign in with Google; new accounts are approved by hand.
Also in L2 The agent loop
- Loop disciplineSettled practice
One plain loop runs the agent: call the model, run the tools it asks for, and repeat until it answers.
- Assistant Must-have
- Coding Must-have
- Research Must-have
- Browser Must-have
- Fresh look each stepSettled practice
Rebuilds what the model sees from the saved record and a fresh look at the world before every step.
- Assistant Must-have
- Coding Must-have
- Research Must-have
- Browser Must-have
- BudgetsOpen problem
Caps on money, time, tokens and repeated actions that stop a run cleanly when one runs out.
- Assistant Must-have
- Coding Must-have
- Research Must-have
- Browser Must-have