OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Memory & RAG

Zengram

Zengram is a self-hosted memory service for multi-agent AI systems, exposed through an MCP server and other documented adapters. It stores typed memories in PostgreSQL and supports retrieval across agents, sessions and machines.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Zengram — OpenViST project profile in Agent Memory & RAG.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Zengram when separate agent sessions need shared, persistent memory rather than isolated conversation context.

Who it is for

  • Developers coordinating multiple AI agents or coding assistants across separate sessions or machines.
  • Teams that want agents to share persistent facts, decisions, status updates or event history.

Documented uses

  • Store a fact discovered by one agent so another can retrieve it later.
  • Search shared memories using semantic or full-text retrieval.
  • Ask for a briefing on updates from other agents.
  • Record structured session reflections through the documented Claude Code skill.
Before you choose

Requirements & limits

The project is open source under the MIT license. The documented setup uses Docker Compose and lists Node.js 20 or later.

  • The memory service uses PostgreSQL, with pgvector for vector search and PostgreSQL full-text search.
  • Entity-graph retrieval and cross-encoder reranking are optional configuration paths.
  • The repository documents an MCP adapter for Claude Code, Cursor and Windsurf; client-specific setup is not provided here for Codex or Gemini.
  • The documented Docker quick start requires a BRAIN_API_KEY and an embedding-provider key.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Start the Zengram service with Docker Compose

Configure and start the documented local service, then check its health endpoint.

Before you start

  • Node.js 20 or later is listed by the repository.
  • Docker Compose
  • A BRAIN_API_KEY and an embedding-provider key

Steps

  1. Clone the Zengram repository and change into its directory.

  2. Copy .env.example to .env and set BRAIN_API_KEY and the embedding-provider key.

  3. Run docker compose up -d.

  4. Check the documented health endpoint at http://localhost:8084/health.

Expected outcome

The service is started through Docker Compose and its health endpoint can be checked.

Guide limitations

  • The guide covers service setup and a health check, not configuring an agent client or verifying memory operations.
  • The repository lists Node.js 20 or later; this guide does not establish compatibility beyond the stated requirement.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
ZenSystemAI/Zengram · Official homepage title · retained in its source language
How we review sourcesSuggest a correction