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

Megamemory

Megamemory is a TypeScript MCP server and CLI that gives coding agents a persistent, per-project graph of concepts, architecture and decisions. It stores the graph in SQLite, supports semantic search with in-process embeddings, and includes a browser-based explorer for viewing concepts and their relationships.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Megamemory — 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 Megamemory when a coding agent needs to carry project-specific concepts and decisions between sessions and retrieve them through MCP. Its graph can also be explored in a browser.

Who it is for

  • Developers using coding agents who need project knowledge to persist across sessions.
  • Teams or individual developers who want to record and query project concepts, architectural decisions and relationships through MCP tools.

Documented uses

  • Orient an agent to a project's top-level concepts at the start of a session.
  • Retrieve relevant architecture or decision context with a natural-language semantic query.
  • Record or update project knowledge after a coding task.
  • Browse a stored knowledge graph and inspect concept details and relationships.
Before you choose

Requirements & limits

The repository is public and lists the MIT license. Installation requires Node.js 18 or later.

  • The agent interprets the code and writes concepts; Megamemory does not use AST parsing or static analysis to index code.
  • Each project uses a SQLite database at .megamemory/knowledge.db by default, with MEGAMEMORY_DB_PATH available to override the location.
  • The documented editor installer supports OpenCode, Claude Code, Antigravity and Codex; other MCP clients can use the stdio server.
  • The embedding model downloads automatically on first use and runs in-process.

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

Install Megamemory for an MCP coding agent

Install the CLI and use its interactive installer to configure a supported editor or MCP client.

Before you start

  • Node.js 18 or later
  • A coding editor with MCP support

Steps

  1. Install the CLI globally with the command below.

    Documented command
    npm install -g megamemory

  2. Run megamemory install and choose an editor in the interactive installer.

  3. For a documented target, run megamemory install --target opencode, claudecode, antigravity, or codex instead.

  4. Restart OpenCode after its target-specific installation; other MCP clients can be configured as a local stdio server running megamemory.

Expected outcome

The selected editor or MCP client is configured to use Megamemory; its project database is stored at `.megamemory/knowledge.db` unless `MEGAMEMORY_DB_PATH` overrides the location.

Guide limitations

  • The embedding model, about 23 MB, downloads automatically on first use.
  • For other MCP clients, the README describes adding a local stdio server but does not provide a universal editor-specific configuration process.
Documentation 1 ↗
Source-based practical guide

Use the documented memory workflow

Orient an agent, retrieve project concepts for a task, and record concepts after work.

Before you start

  • Megamemory configured as an MCP server for the coding agent
  • A project with its Megamemory knowledge database

Steps

  1. At session start, have the agent call list_roots to orient itself.

  2. Before a task, query with understand and a natural-language question, or use get_concept for an exact concept ID.

  3. After work, use create_concept or update_concept to record what was built.

Expected outcome

The agent can retrieve matching graph context and persist new or revised project concepts for later sessions.

Guide limitations

  • The graph stores agent-written concepts, not source-code symbols; the agent reads code and describes it in its own words.
  • The README documents the workflow but does not specify a standard concept-writing format.
Documentation 1 ↗
COMPARE OPTIONS

Compare options for AI coding assistance

These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.

ProjectDocumented purposePlatformAccess and requirements
MegamemoryOrient an agent to a project's top-level concepts at the start of a session.CLI, Self-hosted, ServerThe repository is public and lists the MIT license. Installation requires Node.js 18 or later.
Continuum — pouyahasanamrejiStore project lessons so coding agents can retrieve them across conversations.CLI, Self-hosted, ServerThe repository documents local Docker Compose and bare-metal development setups. Bare-metal startup requires Node 22.12.0 or later and pnpm 10.33 or later; the Compose path skips these host prerequisites.
HaftRecord the problem a project is solving and the options considered.CLI, Self-hosted
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
0xK3vin/MegaMemory · Official homepage title · retained in its source language
How we review sourcesSuggest a correction