Framework-by-framework guides to get cost, reliability and output quality on your agents in production. Most are two lines: install the SDK and point it at your agent.
Install the SDK, attach the LangGraph hook, and every run reports cost, status and quality.
Read the guide →
pythonInstall the SDK, attach the OpenAI Agents SDK hook, and every run reports cost, status and quality.
Read the guide →
pythonInstall the SDK, attach the CrewAI hook, and every run reports cost, status and quality.
Read the guide →
pythonInstall the SDK, attach the Pydantic AI hook, and every run reports cost, status and quality.
Read the guide →
pythonInstall the SDK, attach the LangChain hook, and every run reports cost, status and quality.
Read the guide →
pythonInstall the SDK, attach the LlamaIndex hook, and every run reports cost, status and quality.
Read the guide →
pythonInstall the SDK, attach the AutoGen hook, and every run reports cost, status and quality.
Read the guide →
pythonInstall the SDK, attach the Semantic Kernel hook, and every run reports cost, status and quality.
Read the guide →
typescriptInstall the SDK, attach the Vercel AI SDK hook, and every run reports cost, status and quality.
Read the guide →
typescriptInstall the SDK, attach the Mastra hook, and every run reports cost, status and quality.
Read the guide →
No SDK? Any runtime that can reach a webhook can report a heartbeat. Browse all integrations and the docs.
Free to start, no card. The SDK never blocks or crashes your agents.