DocsInstall

Install

Two lines in your agent, and every model call and tool execution lands on a tamper-evident chain. Start with the verify step if you're evaluating — if a record we produced doesn't survive a command run on your own machine, nothing else here is worth reading.

Two lines

In your agent. Python or TypeScript.

No lines

Through a gateway you already run.

Check it first

Verify a bundle before you send anything.

Python

pip
pip install auditant

import auditant
auditant.init(api_key="ak_…", log_id="tenant_acme/prod", agent_id="underwriter")

With OpenTelemetry present, init()registers a span processor and activates whichever OpenInference instrumentors are installed (LangChain, OpenAI, Anthropic) — capture with no changes to your agent's logic. Install the extras for full coverage:

extras
pip install "auditant[openinference]"     # LangChain / OpenAI / Anthropic auto-capture
pip install "auditant[openai-agents]"     # the OpenAI Agents SDK trace processor
pip install "auditant[gateway]"           # the LiteLLM gateway callback

Without OpenTelemetry, init() still gives you a session on the chain (lifecycle + heartbeats, so a quiet agent is distinguishable from a dead emitter), the manual record() API, and the synchronous decide()policy check. The handle's activated list states exactly what turned on — an audit SDK that implies more coverage than it has would be the product contradiction.

TypeScript — Vercel AI SDK

npm
npm i auditant

import { init } from "auditant";
const auditant = init({ apiKey: "ak_…", logId: "tenant_acme/prod", agentId: "support-bot" });

const result = await generateText({
  model: openai("gpt-4o"),
  experimental_telemetry: { isEnabled: true },   // the AI SDK's own switch
  // …
});
await auditant.flush();   // serverless: before the response returns

The processor attaches to whatever tracer provider your app already runs — this SDK neither installs nor configures OpenTelemetry, because two tracing setups fighting is worse than one. flush()is awaitable and coalesced: a Vercel function freezes the instant the response returns, and an event sitting in a queue at that moment never happened. The AI SDK's own forceFlush routes into it.

What happens next

Events appear on your dashboard's Overview within seconds of the first run. Model calls carry observed token counts (never invented prices — attributionComplete: false until a pricing table says otherwise), tool calls carry input/output hashes, and everything shares a session so the whole arc reads as one record. Next: verify a bundle yourself, or wire an adapter if your agents run behind a gateway or the OpenAI Agents SDK.