Product walkthroughs
Operational, verified walkthroughs for Sentinel, Flight Recorder, Dashboard, Control Center, Registry, SDK and API, what each is, how to use it, and what to expect.
Explore walkthroughsHome / Docs
DocumentationBuild on the evidence layer.
Everything you need to integrate VerifAIer: the evidence engine, twelve governance APIs, one official Python client and the vai command line. All local-first.
Orientation
Get oriented.
What the platform is, what it produced, and a five-minute path through it.
New here? Start with the tour
Get oriented in five minutes, then follow the eleven-stop Guided Platform Tour from install to the API.
Start hereProduct proof
Real captured outputs from the current product, evidence, receipts, passports, reputation, signals and API responses, generated offline with no secrets and checksummed.
See the proofRun it
Get it running.
Install it, and reach your first attested result on your own machine.
First audit in 10 minutes
Install, run your first attested audit, and inspect the receipt, trust score, passport and reputation, offline, no keys.
Start nowInstall
Stand up the evidence engine locally or with Docker in minutes, prerequisites, startup, health check and upgrades.
Install nowQuickstart
From install to your first attested assessment in minutes.
StartReference
Look something up.
The architecture, the internals, and the two programmable surfaces documented exactly as implemented.
Architecture
How the evidence engine, intelligence, identity and distribution layers fit together.
ReadEngineering docs
How VerifAIer works internally, runtime, audit, evidence and trust lifecycles, schemas, provider routing, boundaries and diagrams. Thirty documents, grounded in the implementation.
Read the internalsProduction API docs
The full HTTP API documented exactly as implemented, auth tiers, headers, errors, status codes, idempotency, pagination, rate limits, every endpoint, and curl/Python/TypeScript examples.
Read the API docsAPI reference
Twelve production endpoints for evidence, intelligence, identity and signals.
ReferenceSDK
One official Python client. It prepares a request, or sends it, over the runtime that already exists.
SDK docsProduction
Run it in production.
Configuration, deployment posture and the enterprise rollout, with the known gaps stated rather than omitted.
Operations docs
Run VerifAIer in production, configuration and env reference, local/Docker/compose runbooks, deployment, health, logging, secrets, backup/DR posture, hardening and known gaps. Documented exactly as implemented.
Operate itDeployment
Run local-first, air-gapped or sovereign, the same deterministic engine.
DeployEnterprise guide
Roll out AI trust across teams, fleets and regulated environments.
GuideCommand line
The other runtime.
The vai command line is the same product's developer surface and needs no API at all.
The vai CLI
These docs cover the evidence engine and its API. The vai command-line tool is the same product's developer surface, governance scans, Claude Code hooks, Flight Recorder replays, Memory Receipts, attestation and CI gates, and it is documented in the repository README. It is a separate runtime: it needs no API, and the API needs no CLI.
Quickstart
Your first attested assessment.
# Python from vailidator.sdk import VerifAIerClient client = VerifAIerClient.for_user("you") gov = client.governance.with_base_url("http://localhost:8000").with_sender(send) answer = gov.assess_trust({"evidence": envelope}) print(answer.payload["trust_level"])
Ready to integrate?
Follow the quickstart, or book a working session with our team.