Memorylayer

Hosted memory runtime for agents

Hosted memory runtime
GitHub login Hosted MCP bridge Workspace isolation

Persistent memory for agents, behind one clean service.

Memorylayer turns Engram into something agents can actually connect to: workspaces, API keys, a hosted bridge, bootstrap payloads, and a dashboard that keeps the runtime usable.

Hosted runtime

Long-lived Python service and Postgres, instead of forcing memory through a serverless shape.

Agent bootstrap

Every workspace exposes API URLs, tool discovery, and starter skills for plug-and-play setup.

Free for now

The goal is making hosted memory usable, not gatekeeping it behind fake plans.

Workspace bootstrap API keys, bridge URL, tool discovery, starter skills
HTTP MCP bridge Hosted access to focused Engram tools when local MCP is not an option
Real engine underneath Memorylayer wraps Engram itself instead of replacing it with a fake cloud abstraction

One service around the engine

The core package stays focused on memory behavior. Memorylayer handles the operational surface around it.

Workspace control

Create workspaces, inspect memory health, issue keys, invite collaborators, and keep audit history in one place.

Control
Dashboard, workspaces, invites, API keys, audit feed
Isolation
One Engram schema per workspace with independent memory state

Agent-facing API

Use hosted endpoints for search, remember, recent memories, audit history, bootstrap, and a compact HTTP bridge for selected Engram tools.

Surface
`/bootstrap`, `/status`, `/search`, `/remember`, `/audit`, `/mcp`
Auth
Workspace-scoped keys via `Authorization: Bearer` or `X-API-Key`

Same Engram core

This is still the real Engram engine underneath, so retrieval, graph, and memory improvements pass straight through to the hosted layer.

Runtime
Hosted wrapper around the actual Engram package, not a separate imitation backend
Deploy
VPS-first architecture with a warm Python runtime and direct Postgres access

Made to sit behind agents

The service is shaped for real agent integration, not just for browsing a dashboard.

Bootstrap quickly

Each workspace exposes a bootstrap payload with API URLs, bridge URLs, and starter skills for Codex-style setups.

Use hosted memory directly

Agents can work through the service instead of assuming they have local access to the underlying Engram runtime.

Stay close to the engine

It already has the right shape for dedicated deployments, usage controls, and future agent session capture.