See everything your AI agents

Your agents send emails, create tickets, charge credit cards, and call APIs. AgentLedger logs every action, tracks every cost, and kills agents when things go wrong.

Start Free See live demo
agent.ts
import AgentLedger from 'agentledger';

const ledger = new AgentLedger({
  apiKey: process.env.AGENTLEDGER_KEY
});

// Group related actions into a trace
const traceId = AgentLedger.traceId();

const { result } = await ledger.track({
  agent: 'support-bot',
  service: 'slack',
  action: 'send_message',
  traceId,
  input: { channel: '#support' },
  captureOutput: true,
}, async () => {
  return await slack.chat.postMessage({
    channel: '#support',
    text: 'Issue resolved!'
  });
});

Works with LangChain · OpenAI Agents · CrewAI · MCP Servers · Express · Python & TypeScript

Live Preview

Watch agents in real-time

This is what your dashboard looks like when agents are running. Every action logged, every cost tracked.

live
Waiting for actions...

Features

Everything you need to trust your agents

LLM observability tools track token usage. AgentLedger tracks what agents actually do.

Live Streaming Feed

Real-time SSE streaming of every action. Watch your agents work live with sub-second latency.

Policy Engine & Templates

Define rules or apply pre-built templates: conservative, cost-conscious, compliance. Rate limits, allowlists, cost caps, and more.

Human-in-the-Loop

Require human approval for high-risk actions. Approve or deny from the dashboard, Slack, Discord, or PagerDuty.

Advanced Analytics & Forecasting

Multi-day trend analysis, cost forecasting with linear regression, and budget overrun predictions.

Trace Replay & Debugging

Step through agent traces action-by-action. Inspect input/output at each step with a visual replay timeline.

SDKs, CLI & 8 Integrations

Python & TypeScript SDKs, CLI tool, plus LangChain, OpenAI, MCP, CrewAI, AutoGen, LlamaIndex, and Vercel AI SDK.

Get Started

Three steps. Five minutes.

No infrastructure. No config files. Just wrap and go.

1

Install the SDK

npm install agentledger
2

Wrap your agent actions

await ledger.track({ agent: 'my-bot', service: 'gmail', action: 'send_email', traceId, captureOutput: true, }, sendEmailFn)
3

Watch the dashboard

→ agentledger.co/dashboard

Comparison

Not another LLM tracer

Helicone and Langfuse track LLM calls. We track what happens after.

FeatureAgentLedgerLLM Tracers
LLM call tracing
Real-world action logging
Policy engine & guardrails
Human-in-the-loop approvals
Statistical anomaly detection
Agent evaluations & scoring
Trace replay & debugging
Cost forecasting & analytics
Budget controls & enforcement
Batch logging & data export
CLI tool (npx agentledger)
Slack, Discord & PagerDuty alerts
Live SSE streaming
Python & TypeScript SDKs
8 framework integrations

Open Source

Fully auditable codebase on GitHub

Self-Hostable

Run on your own infrastructure

Fail-Open

Never blocks your production agents

Zero Dependencies

SDK adds no bloat to your project

Pricing

Simple pricing

Start free, scale when you need to.

Available Now

Free

$0forever
  • 5,000 actions/mo
  • 5 agents
  • 7-day data retention
  • Action drawer & I/O
  • Slack, Discord & PagerDuty alerts
  • Community support
Get Started

Pro

$29/month
  • 50,000 actions/mo
  • Unlimited agents
  • 90-day data retention
  • Traces & sessions
  • Budget controls
  • Webhooks
  • Slack, Discord & PagerDuty
  • Email support
Start Pro Trial

Team

$99/month
  • 500,000 actions/mo
  • Unlimited agents
  • 1-year data retention
  • Traces & sessions
  • Budget controls
  • Webhooks
  • Slack, Discord & PagerDuty
  • SSO & SAML
  • Priority support
Start Team Trial

Free tier includes 5,000 actions/month with 7-day data retention. Usage beyond plan limits is rate-limited. We reserve the right to enforce fair use policies to maintain service quality for all users.

FAQ

Common questions

Stop flying blind with your agents

Know exactly what your agents are doing, what they cost, and how to stop them when things go sideways.

Get Started Free