Platform & API

Built for the agentic future

Arvanic isn't just a product you log into — it's a platform your own AI agents can work alongside. A scoped read API, a native Model Context Protocol server, and signed webhooks let you automate the busywork around an investigation while the sensitive, human parts stay exactly where they belong: with you.

Start free Read the developer docs How we handle data

Three ways to connect

Read your data, let an agent operate it, and get pushed the moment something happens — all under the same keys and the same confidentiality controls.

Read API

A clean, versioned REST API

Query investigations, participant progress, the question plan, and generated reports over a stable /api/v1 surface. Authenticated with scoped API keys you create and revoke yourself.

List & inspect investigations
Participant progress (role-labeled)
Download generated reports

MCP server

Native Model Context Protocol

Point any MCP-capable AI agent at Arvanic and it discovers what it can do automatically — no custom integration. Your assistant can answer “what’s the status of my open investigations?” out of the box.

Auto-discovered tools
Same scoped-key auth
Works with modern AI agents

Webhooks

Real-time, signed event delivery

Get notified at your own endpoint the moment a report is generated, a participant submits, or an investigation closes — every delivery cryptographically signed so you can trust it came from Arvanic.

report.generated
interview.submitted
investigation.closed

Up and running in three steps

Create a scoped key

In Settings → API keys, generate a read-only key with exactly the permissions you want. Copy it once.

Connect your agent

Point an MCP client at the Arvanic server with your key, or call the REST API directly. Tools are discovered automatically.

Automate & get notified

Your agent reads what it needs; webhooks push the events you care about to your systems in real time.

Secure by design

Automation without surrendering confidentiality

Workplace investigations are sensitive and legally consequential. Every part of the platform is built so that opening Arvanic to your agents never means opening it to risk.

Scoped, revocable keys

Every key carries only the permissions you grant (read-only today) and can be revoked instantly. Secrets are stored hashed and shown to you once.

Confidentiality by default

Programmatic reads are de-identified: participants come back role-labeled (“Complainant 1”), never as names, unless you explicitly elevate access. The same PII protections that govern the product govern the API.

Tenant isolation

Keys are bound to your organization. There is no path to another organization’s data — enforced at the database layer, not just the application.

Human stays accountable

Agents read, prepare, and notify. The legally-significant actions — sending invitations, finalizing reports, certifications — remain with the named human investigator. An agent acts on your behalf; it never becomes you.

Signed webhook payloads

Each delivery includes an HMAC signature and carries minimal, non-identifying data — your systems verify authenticity and fetch detail under your own key.

Audited & rate-limited

Key creation and revocation are recorded in the immutable audit trail, and every programmatic surface is rate-limited and abuse-protected.

Where this is going

An investment in future value

The foundation is live today, and it's deliberately built to grow. As AI agents become part of how teams work, Arvanic is already the system of record they can plug into — securely, and on your terms.

Available now

Read, connect, get notified

Scoped API keys, the REST read API, the MCP server, and signed webhooks — live today in your account settings.

Agent-prepared actions, human-approved

Let an agent draft a question plan or stage participants, then approve it with one tap. The agent does the legwork; you make the call.

On the roadmap

Live progress streaming & deeper integrations

Streaming progress for long-running operations and richer connectors as the agentic ecosystem matures — built on the same secure, auditable foundation.

Bring your agents to your investigations

Create an account, generate a key in Settings, and connect in minutes. Questions about a specific integration? Talk to us.

Get Started