Name the three canonical agent stopping conditions and explain why cost cap is the hardest to implement correctly

Attempt it

What are the three canonical stopping conditions for an agent loop? For each, briefly describe how it is detected. Then explain why cost cap is harder to implement correctly than the other two.

Free · 2 AI evals / day

Quick reference

TL;DR

The three stopping conditions are final-answer signal, step budget, and cost cap. Cost cap is hardest because next-step cost is unknown until after the call and compounds as context grows.

Memory aid

Sign in for the mnemonic that makes this stick. See Pro pricing.

Easy to grasp

Imagine you give a helper a task and a few rules for stopping. First, they stop when they say they are finished. Second, they stop after a fixed number of tries, no matter what. Third, they stop once they have spent a set amount of money. The first two rules are easy to check. "Did they say done?" is a yes or no question. "Have they tried ten times?" is just counting. But the money rule is sneaky. You only learn the price of a try after it happens, not before. And each new try costs more than the last, because the helper carries a growing pile of notes along every time. So it is easy to blow past your budget on the very step that pushes you over, since you could not see the bill until the work was already done.

Key concepts

Stopping Conditions Cost Budgeting Step Budget

Detailed answer & concept explanation

~8 min read

Everything you need to truly understand this topic: intuition, mechanics, step by step explanation, code, formulas, and worked example. Click to expand.

Everything you need to truly understand this topic: intuition, mechanics, step by step explanation, code, formulas, and worked example.

Sign in for 5 free unlocks/day or go Pro for unlimited. See Pro pricing.

Quick scan · 2–4 min

Everything important, quickly.

Name the three stops and their detection mechanisms, frame steps and cost as hard backstops layered under the model's own done signal, then spend most of the answer on why cost is hardest: post hoc knowledge of token cost, monotonic compounding of context, overshoot risk, and what to return on a forced stop.

Real-world examples

Real products, models, and research that use this idea.

LangGraph exposes a recursion_limit that caps loop iterations and raises a GraphRecursionError, the framework-level step budget acting as a hard backstop.
The Anthropic SDK returns a usage field with input and output token counts on every call, letting a loop reconcile real cost after each step rather than guessing.
OpenAI's Assistants and Responses APIs expose run-level step limits and token accounting so an agent can be cut off before it drains a project's spend.
Cursor and Cline coding agents cap edit-test iterations and surface a token or request budget, halting the loop and returning partial diffs when the ceiling is hit.

Interviewer follow-ups

What an interviewer would ask next. Try answering before peeking at the approach.

QHow would you enforce a cost cap that almost never overshoots, even on a step that returns a huge tool output?

Sign in or go Pro to see the approach for this follow-up.

QWhy is verifying the model's final-answer signal safer than trusting it outright, and how would you verify it?

Sign in or go Pro to see the approach for this follow-up.

QWhat should the loop return when a step or cost backstop forces a stop mid-task?

Sign in or go Pro to see the approach for this follow-up.

Don't say this

Red flags and common mistakes that signal junior thinking. Click to expand.

Most common mistake

Trusting the model's final-answer signal as the only stop. A confused agent declares done prematurely or never, so hard backstops on steps and cost must run independently of what the model claims.

Sign in or go Pro to see all red flags and common mistakes. See Pro pricing.

60-second cheat sheet

The night-before-the-interview bullets. Scan these on the way to the call.

Sign in for 5 unlocks/day or go Pro for unlimited revision sheets. See Pro pricing.

References

Primary sources. Skim if you want the original framing.