Skip to main content

​
Cost Budget Auto-Stop

The CostTracker now checks budgets after every tool roundtrip, not just before the loop starts. This prevents a multi-roundtrip agent from blowing through its budget.

​
How It Works

When a costTracker is configured on an agent, the onRoundtripComplete loop hook automatically:
  1. Tracks incremental token usage after each roundtrip
  2. Checks if any budget limit is exceeded
  3. If exceeded, gracefully stops the LLM loop and returns the last assistant message

​
Configuration

import { Agent, openai, CostTracker } from "@radaros/core";

const tracker = new CostTracker({
  pricing: { "gpt-4o": { promptPer1k: 0.0025, completionPer1k: 0.01 } },
  budget: {
    maxCostPerRun: 0.50,     // $0.50 max per run
    maxTokensPerRun: 100_000, // 100K tokens max per run
  },
});

const agent = new Agent({
  name: "budget-bot",
  model: openai("gpt-4o"),
  costTracker: tracker,
});

​
New CostTracker Methods

// Track mid-loop usage
tracker.trackIncremental({ runId, agentName, modelId, usage });

// Check without throwing
const exceeded = tracker.isBudgetExceeded(runId, sessionId, userId);

// See remaining budget
const { costRemaining, tokensRemaining } = tracker.estimateRemaining(runId);

​
Behavior

ScenarioBeforeAfter
10-roundtrip loop, budget hit at roundtrip 3Budget only checked before loopLoop stops at roundtrip 3
Cost trackingOnly after full loop completesTracked incrementally each roundtrip