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

Engram — 199-biotechnologies

Engram is a local-first MCP server that gives compatible AI clients persistent memory across conversations. It provides tools to store, recall and forget memories, combining BM25 keyword search, semantic embeddings and knowledge-graph lookup, then ranking results with temporal decay and salience.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Engram — 199-biotechnologies — 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 Engram when an MCP-compatible assistant needs searchable context that persists beyond a single conversation, especially when keyword matches, semantic similarity and relationships between entities are all useful.

Who it is for

  • Developers connecting an MCP-compatible AI assistant to persistent memory.
  • Claude and Claude Code users who want context to carry between conversations.
  • Teams or builders exploring hybrid retrieval and linked memory entities in a local-first MCP service.

Documented uses

  • Store personal or project context so an assistant can retrieve it in later conversations.
  • Find relevant memories by exact terms as well as conceptually related wording.
  • Surface connected details about people, projects and preferences through entity relationships.
  • Remove a specific stored memory when it is no longer wanted.
Before you choose

Requirements & limits

The repository is public and identifies the project as MIT-licensed. Setup requires Node.js 18 or later. The README's Claude Desktop example uses an Anthropic API key for that client configuration; the key is listed as optional and enables memory consolidation.

  • Engram is an MCP server to connect to an AI client, not a standalone conversational assistant.
  • Hybrid retrieval fuses BM25, semantic embeddings and knowledge-graph lookup before temporal and salience scoring.
  • The default transport is stdio; HTTP transport and a configurable port are also documented.
  • An Anthropic API key enables memory consolidation, but is listed as optional.

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

Connect Engram to Claude Code

Register Engram as an MCP server for Claude Code.

Before you start

  • Node.js 18 or later
  • Claude Code

Steps

  1. Install Engram globally with the command below.

    Documented command
    npm install -g @199-bio/engram

  2. Register it with Claude Code using claude mcp add engram -- npx -y @199-bio/engram.

Expected outcome

Claude Code has an Engram MCP server entry for persistent memory access.

Guide limitations

  • The instructions describe registration; they do not document a verification or troubleshooting procedure.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
16 October 2026
Original website title
199-biotechnologies/engram · Official homepage title · retained in its source language
How we review sourcesSuggest a correction