Audifact Beta
Append-only event ledger for AI agents

What agents knew at decision time

Audifact records decisions, context, and outcomes in a hash-chained ledger so you can prove the story later.

One .wrap() on LangGraph, CrewAI, and friends — or direct API calls.

Enterprise & compliance Startups & product teams Weekend builders

No credit card · Demo mode works without an API key · Free tier to start

Public beta — product is still stabilizing. Sign up free, try the quickstart, and tell us what breaks.

WORKS WITH

Drop-in tracers for the stacks you already run. Same decision → outcome model, same Explorer — whether you're on LangChain, LlamaIndex, or multi-agent frameworks.

LangChain & LlamaIndex tracers · LangGraph / CrewAI / AutoGen via AudifactTracer.wrap()

WHY AUDIFACT

Decision-time context,
sealed when it matters

Not reconstructed from logs days later. Attach outcomes and reopen the chain with proof.

One line to start

tracer.wrap(...) on the frameworks you already run — or direct API calls. No logging pipeline, no custom middleware.

🔒

Privacy you control

You control what's exposed — optional public_decision_label / public_outcome_label (and structured public JSON) on proof links. Turn on include_context=True only when you need full inputs, docs, or tool outputs.

🔗

Decisions, outcomes when you need them

Log decisions on their own. Attach outcomes later when results exist — linked in the Explorer with readable labels, not a wall of JSON.

REAL EXAMPLES

Pick your starting point

Runnable demos, copy-paste snippets, and use-case guides — from enterprise workflows to weekend projects.

Start small. Scale when it matters.

Run some examples locally in demo mode tonight. Add an API key when you're ready to persist signed events to the cloud.

Also explore CrewAI and AutoGen enterprise guides.