L2 The agent loop
Budgets
Caps on money, time, tokens and repeated actions that stop a run cleanly when one runs out.
Why it matters
Without enforced caps an agent can spend without limit, run past its deadline or repeat the same action forever. A clean stop also hands back the work done so far.
What to build
One governor checks every cap each step
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
- 31
- Checked facts behind it
- 379
See the 379 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
- StoppingSettled practice
Ends a run in a clear, named way, hands back the work done, and checks that "done" really means done.
- Assistant Must-have
- Coding Must-have
- Research Must-have
- Browser Must-have