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

Mengram

Mengram is an open-source developer tool that gives AI-agent applications semantic, episodic and procedural memory through an API, Python and JavaScript SDKs, and named integrations with LangChain, CrewAI and OpenClaw.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Mengram — 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

Consider Mengram when an agent needs more than in-session context and should retrieve facts, past events or workflows. The README also offers a local-only preview command for exploring memory before connecting through the API.

Who it is for

  • Developers building AI agents that need persistent facts, event history or evolving workflows.
  • Teams using LangChain, CrewAI or OpenClaw that are evaluating a named memory integration.

Documented uses

  • Add stored facts and conversation information to an agent application.
  • Search remembered facts or ask for a synthesized answer with citations.
  • Retrieve past episodes or procedural workflows, including procedures described as learning from failures.
Before you choose

Requirements & limits

The project advertises a free API key. The README directs users to Mengram’s website to obtain a key for SDK use; its `mengram try` preview is described as local-only and requires no account.

  • The repository is public and licensed under Apache-2.0.
  • The README demonstrates add, search, ask, episode-query and procedure-query operations.
  • The SDK examples use an API key; the local preview is a separate, no-account path.
  • Integration names are documented, but the supplied material does not detail or verify their implementation.

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

Add the Mengram connector to Claude

Connect Claude to Mengram so it can save and recall memory across conversations.

Before you start

  • Claude access to Settings → Connectors.
  • An email address for Mengram sign-in.

Steps

  1. In Claude, open Settings → Connectors and choose Add custom connector.

  2. Enter the name Mengram and server URL https://mengram.io/mcp/connector.

  3. Leave the OAuth Client ID and Secret fields empty, then choose Add.

  4. Complete the Mengram sign-in using your email and the code sent to you.

Expected outcome

The connector is live, and Claude can use memory tools to remember, recall, load context, and list procedures.

Guide limitations

  • Free Claude accounts can add one custom connector.
  • This connector path applies to Claude on claude.ai, Desktop, and Cowork; Claude Code uses a separate hooks integration.
Documentation 1 ↗
Source-based practical guide

Install the Mengram Python SDK and try local memory preview

Install the Python SDK and preview what Mengram memory would know locally.

Before you start

  • Python package installer pip.

Steps

  1. Install the SDK with the command below.

    Documented command
    pip install mengram-ai

  2. Run mengram try to see what memory would know about you.

Expected outcome

The SDK is installed and the preview runs locally without an account or sending data off the machine.

Guide limitations

  • This preview is local-only; these steps do not configure a connected agent or demonstrate remote memory use.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗Guide documentation ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
alibaizhanov/mengram · Official homepage title · retained in its source language
How we review sourcesSuggest a correction