Governed memory for AI.Inspect. Edit. Delete. Control.
A governance layer for AI memory. Store, audit, scope, and expire what AI systems remember — with full provenance.
Governed recall, with provenance
GOVERNANCE FEATURES
How It Works
Three simple steps to give your AI governed, explicit memory.
Connect Sovant
Add the SDK or MCP plugin. No lock-in. Memory stays explicit and controlled.
Write memory explicitly
Persist facts, decisions, and preferences intentionally — not auto-captured guesswork.
Recall with scope and provenance
Retrieve the right context with clear source, scope, and lifetime.
Why Governance?
Most AI memory is invisible, noisy, and permanent. Sovant is different.
Explicit, not auto-captured
You decide what gets stored. No guessing, no noise, no surprises.
Inspectable, not hidden
Every memory is visible. Edit, review, or delete it — anytime.
Controlled, not permanent
Scope by project, set expiry, and revoke access. Your data, your rules.
See Governed Memory in Action
Inspect, edit, and control what AI uses — nothing is implicit.
The Dashboard
A single surface to monitor, chat, and govern every memory.

Workspace & Memory Activity
Monitor, manage and track your project's memory activity in one place.

Workspace Chat (governed recall)
Conversations with governed recall — every response grounded in auditable memory.

Memory Governance
Inspect, edit, delete, and audit memories with full provenance.
Built for Developers
A governed API that makes AI memory explicit, inspectable, and controllable.
Governance-first: Edit, delete, expire (TTL), and scope memory with full provenance.
Every memory is auditable — know where it came from, who wrote it, and when it expires.
Scoped recall: Retrieve the right context with clear source and lifetime.
Memories are scoped per project, thread, or globally — never silently expanded.
Type-safe SDKs: Full TypeScript and Python support with auto-completion.
Build faster with intellisense and type safety.
npm i @sovant/sdk
// index.ts
import { Sovant } from '@sovant/sdk';
const client = new Sovant({
apiKey: process.env.NEXT_PUBLIC_SOVANT_API_KEY!,
baseUrl: 'https://sovant.ai',
});
// Create a memory
await client.memory.create({
content: 'User prefers dark mode UI',
type: 'preference',
tags: ['ui', 'theme'],
});
// Semantic search
const results = await client.memory.search({
query: 'user theme preference',
limit: 5,
});
console.log(results);Simple, Transparent Pricing
Start free, scale as you grow
Free
Full access to governed memory — no credit card required
- 2,000 requests per month
- Governed memory core
- Dashboard & Memory Governance
- SDK + MCP plugin access
- Community support
Founding Developer
Support Sovant's early development
- Early supporter badge
- Priority feedback channel
- First access to Pro features
- Recognition in community page
- Optional upgrade (no feature gating)
Enterprise & Compliance
Advanced controls, compliance workflows, and custom deployments
- Higher rate limits and SLAs
- Role-based access controls
- Custom deployment options
- Dedicated support
A 'request' counts any API call that reads, writes, or patches memory.