Skip to content
agentFast
Documentation

The production framework
for AI agents.

CrewAI and LangGraph frame how your agent thinks. agentFast frames everything above that — and stays agnostic about which one runs underneath. This is what a run looks like once it does:

run trace
planningcomplexitymedium — budget raised to 30 iterations
memoryrecall2 facts about cust_alice
guardrailpii:input1 email redacted before the model saw it
llmclaude-sonnet-4-51 tool call$0.0031
toolkb_searchok · 3 hits · 6ms
llmclaude-sonnet-4-51 tool call$0.0028
toolorder_lookupok · ord_1001 · 4ms
hitlapproval_requestedrefund_request — run paused, durably
A support agent answering a refund ticket. It sized up the task, recalled what it knew, searched the KB, looked up the order — then stopped, because refunds are high-risk and a human has to say yes. Kill the process here and the run survives; approve later and it continues from this exact step.

Not a developer? Start here — no code on that page. Every capability below also opens with a plain-language summary before the technical detail.

Get started

From a clean machine to an agent answering a real ticket.

Build

Make it yours: your tools, your behaviour, your infrastructure.

The production layer

The capabilities agentFast frames — identical on every SDK.

Reference

Every command, endpoint, and configuration key.