# Recall

Recall is an MCP server that provides self-hosted, encrypted user memory shared across AI assistants. Runs locally without external network dependencies during execution.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 5 repo stars
- Author: j4ckxyz
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/j4ckxyz/recall
- Updated: 2026-10-23
- Security check: automated, 2026-11-17 (text) — warning: a command downloads a script from github.com and runs it right away; a PowerShell command downloads a script from github.com and runs it right away. Do not run such commands without an explicit yes from the human.
- HTML: https://affpapa.org/en/nevibecode/mcp/j4ckxyz-recall

## What it does

Recall runs as a self-contained 10 MB binary with a built-in embedding model, operating without Python, GPUs, or runtime network requests. Memory storage relies on AES-256 encryption via SQLCipher on your drive. Agents interact with the store through three dedicated MCP actions: `recall`, `remember`, and `forget`.

Information retrieval pairs BM25 keyword matching with static embeddings and spelling correction, executing lookups in roughly 150 microseconds. An automated scanner checks every write operation to intercept leaked credentials or sensitive IDs. A lightweight 65 KB password-guarded web UI is included inside the same binary for auditing and modifying stored records.

Decryption depends entirely on `~/.recall/key`, so losing that file makes the database unrecoverable. Integration with password vaults is handled through the `RECALL_KEY_CMD` environment variable.

## Install

### Claude Code (native platform)

**Install**
- The MCP registry has no launch command for it: see the repository README

## When to use it
- share user context across multiple independent AI clients
- keep personal instructions on local disk under AES-256 encryption
- prevent credentials and secret keys from entering agent memory
- inspect and modify agent-stored records inside a local web UI

## Example prompts
- Remember that I deploy staging environments with Podman instead of Docker
- Find my preferred commit message conventions with recall
- Forget my previous office IP address from memory

## Similar
- [SQLCipher Encrypted Database Expert](https://affpapa.org/en/nevibecode/skills/sqlcipher-encrypted-database-expert.md) — skill · 216 installs
- [mempalace-recall](https://affpapa.org/en/nevibecode/skills/mempalace-recall.md) — skill · 2000 installs
- [gsd-mempalace-recall](https://affpapa.org/en/nevibecode/skills/gsd-mempalace-recall.md) — skill · 573 installs
- [Compartment](https://affpapa.org/en/nevibecode/mcp/compartment.md) — MCP server · ★ 472 repo stars
- [Synapse Layer — Trust Infrastructure for AI Agents](https://affpapa.org/en/nevibecode/mcp/synapse-layer-trust-infrastructure-for-ai-agents.md) — MCP server · ★ 13 repo stars
- [OwnMemory](https://affpapa.org/en/nevibecode/mcp/ownmemory.md) — MCP server · ★ 1 repo stars

## Other versions with the same name
- [Recall · polign](https://affpapa.org/en/nevibecode/mcp/polign-recall.md) — ★ 16 repo stars
- [recall · recallworks](https://affpapa.org/en/nevibecode/mcp/recall.md) — ★ 8 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
