# OMEGA Memory

OMEGA Memory — an MCP server that gives a coding agent long-term memory for decisions, mistakes and task checkpoints, found later by meaning or through the graph.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · 2862 installs
- Author: omega-memory
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://smithery.ai/server/omega-memory/core
- Updated: 2026-02-15
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/omega-memory-omega-memory

## What it does

Everything sits on a remote endpoint at `https://core--omega-memory.run.tools`, so notes outlive a restart. Types cover decisions, lessons, errors, preferences and checkpoints. Registering it in Claude Code takes one `claude mcp add` line.

The tools carry an `omega_` prefix. `omega_store` writes, `omega_query` searches by meaning, `omega_traverse` walks relations from a starting memory, `omega_compact` and `omega_consolidate` merge near-duplicates and drop dead weight. Long jobs get `omega_checkpoint` and `omega_resume_task`: plan, progress, touched files and choices are captured, then loaded back in the next session. Reminders set with `omega_remind` arrive when due, either at session start or mid-work.

Mind the query mode. `omega_query` defaults to semantic matching, so an exact error string needs `mode='phrase'`; otherwise a literal match can slip away. A local route exists too: `pip install omega-memory && omega setup`. The author points to first place on LongMemEval at 95.4%.

## Requirements
- Remote server: `https://core--omega-memory.run.tools`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http core https://core--omega-memory.run.tools
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add omega-memory-omega-memory --url https://core--omega-memory.run.tools
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "omega-memory-omega-memory": {
            "url": "https://core--omega-memory.run.tools"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "omega-memory-omega-memory": {
            "type": "http",
            "url": "https://core--omega-memory.run.tools"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http omega-memory-omega-memory https://core--omega-memory.run.tools
```

## When to use it
- restore a task's context after the session restarts
- find an earlier decision by meaning, not by exact wording
- checkpoint progress before the context window fills up
- prune stale notes and merge near-duplicate ones
- get a due reminder at the start of the next session

## Example prompts
- Remember that we dropped the Redis cache because invalidation broke the tests — save it with `omega_store`.
- Pull everything we decided about the migrations and show me the timeline by day.
- Checkpoint this task now, then bring it back tomorrow with `omega_resume_task`.

## Similar
- [memory-search](https://affpapa.org/en/nevibecode/skills/ruvnet-memory-search.md) — skill · 751 installs
- [Omega](https://affpapa.org/en/nevibecode/mcp/omega.md) — MCP server · 3525 installs
- [Memory Tool](https://affpapa.org/en/nevibecode/mcp/memory-tool.md) — MCP server · 107 installs
- [openviking-memory](https://affpapa.org/en/nevibecode/skills/openviking-memory.md) — skill · 545 installs
- [shodh-memory](https://affpapa.org/en/nevibecode/mcp/shodh-memory.md) — MCP server · ★ 322 repo stars
- [hot-memory-mcp](https://affpapa.org/en/nevibecode/mcp/hot-memory-mcp.md) — MCP server · ★ 15 repo stars

## Other versions with the same name
- [OMEGA Memory · omega-memory](https://affpapa.org/en/nevibecode/mcp/omega-memory.md) — ★ 226 repo stars

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
