Culvii Kit · Documentation

From zero to a governed agent in 5 minutes.

Everything you need to build enterprise AI agents with cryptographic identity, capability governance, HITL controls, and immutable audit trails.

From zero to a governed agent in 5 minutes.

Prerequisites: Node.js 18+ or Python 3.11+. A Culvii account — start your 30-day trial.

1

Install

npm pip
npm install @culvii/kit
2

Initialise your tenant

npx culvii init
# Creates .culvii/ and provisions your API key
3

Register your first agent

import { Actor, Model } from '@culvii/kit'

const agent = new Actor({
  name: 'my-first-agent',
  model: Model.openai('gpt-4o'),
  capabilities: ['read:documents'],
})

console.log(agent.did)
// → did:culvii:my-tenant:my-first-agent-a3f9b2...
4

Run it

const result = await agent.run({
  prompt: 'Summarize the Q1 financial report.',
  context: { document: '...' },
})
// Every step is signed, hashed, and logged
5

Verify the audit trail

import { Auditor } from '@culvii/kit'

const proof = await Auditor.verify({
  agentDid: agent.did
})
// → { valid: true, events: 3,
//     merkleRoot: 'a3f9b2...', tampered: false }
Documentation structure

Five primitives. Infinite agents.

Each primitive maps to a real engineering concern — not an academic abstraction. They compose.

Five ideas that underpin everything.

Agents are first-class citizens

Agents aren't just function calls or prompt chains. They're named, identity-bearing entities registered in your tenant — accountable in the same way a service account is accountable.

Capabilities are explicit, not ambient

Agents can only do what they've been explicitly granted. No capability, no execution — regardless of what the LLM is capable of. Capabilities expire, rotate, and revoke instantly.

Every action is observable and immutable

You don't need to add logging. Every tool call, model invocation, and workflow step emits an observation event that flows through a hash chain anchored to a Merkle Tree root.

Workflows are governed, not just orchestrated

A Culvii Workflow is a governed execution context with checkpoints, HITL gates, and rollback semantics. Pause execution, require approval, and resume — without losing state.

Tenants are the unit of isolation

Everything — agents, capabilities, keys, logs — is scoped to a tenant. Tenants never share data, keys, or audit logs. Full isolation at the database, application, and cryptographic layers.

Need help?

Join our Discord for questions, or open a GitHub issue for bugs and feature requests.