# Agent Pulse docs

Record what each model and tool call costs, and see it by agent, customer, user and model.

## Before you start {#before-you-start}

- **An organisation**: Sign up and create one. It holds your agents, keys and team.
- **An API key with Write**: Create it on the API keys page. The secret is shown once.

## Quickstarts {#quickstarts}

Pick how your agent calls models. Each ends with your agent on the Agents page.

- [Python](/docs/quickstart): OpenAI, Anthropic and Gemini clients, ADK, LangChain
- [TypeScript](/docs/quickstart): Vercel AI SDK, or recording by hand
- [Go](/docs/quickstart): Gemini, Anthropic and OpenAI clients, Google ADK
- [LiteLLM proxy](/docs/litellm): One line of config records every service behind it
- [OpenTelemetry](/docs/opentelemetry): Send the GenAI spans you already emit, from any language

## Guides {#featured-guides}

- [Track cost per customer and user](/docs/cost-per-customer): Say who each request is for.
- [Set a budget](/docs/budgets): Alert, switch to a cheaper model, or stop calls at a limit.
- [Find what is failing](/docs/failures): What failed calls and retries cost, and where.
- [Read your spend over the API](/docs/reading-spend-api): Put the figures in your own screens.

## What it is for {#background}

- [AI agent metering](/ai-agent-metering): What to record about each agent call, and what it lets you do.
- [LLM cost tracking](/llm-cost-tracking): How a call is priced, and how to read cost by customer and request.

## How it compares {#compare}

- [Agent Pulse vs Langfuse](/agent-pulse-vs-langfuse): Cost by customer with budgets, beside tracing and evaluation.
- [Agent Pulse vs Helicone](/agent-pulse-vs-helicone): An in-process recorder, beside an AI gateway.
- [Agent Pulse vs LiteLLM spend tracking](/agent-pulse-vs-litellm): The proxy's own spend tracking, and what Agent Pulse adds.

## Every page {#every-page}

### Start {#start}

- [Quickstart](/docs/quickstart): Record your agent's first model call and see it in the console.
- [Assess your codebase](/docs/assess): A prompt for your coding agent that reads your repository and reports what Agent Pulse would show you and what it would take to connect.
- [Ask your AI assistant](/docs/ai-assistants): Connect Claude, ChatGPT, Cursor or VS Code to Agent Pulse and ask about your agents' spend.

### Guides {#guides}

- [Track cost per customer and user](/docs/cost-per-customer): Tag each request with the customer and the signed-in user.
- [Set a budget that warns or stops calls](/docs/budgets): Alert at a share of a limit, and either stop covered model calls at the limit or only warn.
- [Get alerts in Slack or a webhook](/docs/alerts): Which alerts are raised, and how to send them to Slack or your own endpoint.
- [Find what is failing](/docs/failures): What failed calls and retries cost, which agents are failing, and why.
- [Compare cheaper models](/docs/cheaper-models): See what an agent's own calls would have cost on other models that can do the same work.
- [Read your spend over the API](/docs/reading-spend-api): Put your organisation's figures in your own screens, a warehouse or a report.
- [Check your bill](/docs/bill-check): Send your Google bill once a day and see each hour checked against what was recorded.

### Frameworks {#frameworks}

- [OpenAI, Anthropic and Gemini clients](/docs/model-clients): Record every call a model client makes by wrapping it once.
- [Google ADK](/docs/google-adk): Register the recorder once on an ADK agent; no call site changes.
- [LangChain and LangGraph](/docs/langchain): One handler on the outermost call records every model and tool call in the run.
- [OpenAI Agents SDK](/docs/openai-agents): Run hooks record every model and tool call, under the agent that made it.
- [Vercel AI SDK](/docs/vercel-ai-sdk): A telemetry integration records every call; the middleware lets a budget stop one.
- [LiteLLM proxy](/docs/litellm): One line in the proxy's config records every model and service behind it.
- [OpenTelemetry](/docs/opentelemetry): Point your OTLP exporter at Agent Pulse and send the GenAI spans you already emit.
- [Pydantic AI](/docs/pydantic-ai): Send Pydantic AI's own OpenTelemetry spans to Agent Pulse.
- [Strands Agents](/docs/strands-agents): Send Strands' own OpenTelemetry spans to Agent Pulse.
- [Semantic Kernel](/docs/semantic-kernel): Send Semantic Kernel's own OpenTelemetry spans to Agent Pulse.
- [Microsoft Agent Framework](/docs/microsoft-agent-framework): Send Agent Framework's own OpenTelemetry spans to Agent Pulse.
- [Mastra](/docs/mastra): Send Mastra's own OpenTelemetry spans to Agent Pulse.
- [OpenLLMetry](/docs/openllmetry): Record any Python framework built on the OpenAI client with Traceloop's OpenLLMetry.
- [OpenLIT](/docs/openlit): Record any Python framework built on the OpenAI client with OpenLIT.
- [CrewAI](/docs/crewai): Record CrewAI's model calls with OpenLLMetry or OpenLIT.
- [LlamaIndex](/docs/llamaindex): Record LlamaIndex's model calls with OpenLLMetry.
- [Haystack](/docs/haystack): Record Haystack's model calls with OpenLLMetry.
- [AutoGen](/docs/autogen): Record AutoGen's tool calls from its own spans and its model calls with OpenLLMetry.

### Concepts {#concepts}

- [What is recorded, and what never is](/docs/what-is-recorded): The fields on each record, and the content that never leaves your process.
- [How cost is calculated](/docs/how-cost-is-calculated): Priced on the server from token counts, at the rate in force when the call was made.
- [Organisations, workspaces and requests](/docs/requests-and-workspaces): How spend is grouped, from your organisation down to one end-user request.
- [When Agent Pulse is unavailable](/docs/when-agent-pulse-is-unavailable): The recorder never slows or breaks your agent; what happens when we are down.
- [Security and trust](/docs/security): What never leaves your process, what happens when we are down, and how keys and records are kept.

### API reference {#api}

- [API overview](/docs/api-overview): Base address, keys, money, paging and the OpenAPI document.
- [List activities](/docs/api-list-activities): Individual model and tool calls, newest first.
- [Aggregate spend](/docs/api-aggregate-spend): Spend grouped by agent, customer, user, model, day and more.
- [List agents](/docs/api-list-agents): The agents recording under your organisation.
- [List users](/docs/api-list-users): The names your agents have sent for user identifiers.
- [List outcomes](/docs/api-list-outcomes): Units of business work counted against a request.
- [List prices](/docs/api-list-prices): The rate card every call is priced against.
- [Report a bill](/docs/api-report-bill): Send a day of your Google bill and get each hour and model compared.

### More {#more}

- [Recorder settings](/docs/recorder-settings): The environment settings every recorder reads, and the options you can tune.
- [Error codes](/docs/error-codes): Failure kinds on recorded calls, and the errors the API and recorders return.
- [Changelog](/docs/changelog): How the API changes, and recorder releases, newest first.