RecallNest
Local-first shared memory layer for AI clients with hybrid recall and session continuity.
Tested on LinuxKnown vulnerabilities
Use it
Before you install: known vulnerabilities
pickfit found 1 known high or critical vulnerability in the packages a fresh install brings (OSV.dev, 3 Nov 2026). Weigh this before you install it.
npm:zod 3.25.76Zod can exhaust memory while validating oversized constrained collections (GHSA-mrhx-c7xc-m3w5, high, fixed in 4.6.6)
Linux x86_64
You need first
- Runtimes:
Bun - Keys or accounts: JINA_API_KEY
download Bun installer
curl -fsSL https://bun.sh/install -o /tmp/bun-install.shinstall Bun prerequisite
bash /tmp/bun-install.sh && export PATH="$HOME/.bun/bin:$PATH"npm install -g recallnest
export PATH="$HOME/.bun/bin:$HOME/.local/bin:$PATH" && npm install -g recallnest
Where the README and reality differ
- README: npm install -g recallnest works via Node
What worked: requires Bun on PATH first; launcher exits with instructions if bun missing - README: MCP server starts after install
What worked: config.json not found until you copy config.json.example to ~/.config/recallnest/config.json
Add it to your agent
claude mcp add recallnest -e JINA_API_KEY='<JINA_API_KEY>' -- bun run '<install-folder>/src/mcp-server.ts'JSON for Claude Desktop, Cursor and other clients
{
"mcpServers": {
"recallnest": {
"command": "bun",
"args": [
"run",
"<install-folder>/src/mcp-server.ts"
],
"env": {
"JINA_API_KEY": "<JINA_API_KEY>"
}
}
}
}You fill in
JINA_API_KEY: embeddings API key (required)
Replace <install-folder> with the folder you ran the install steps in. Ask the user for JINA_API_KEY; pickfit never had a value, so these are placeholders.
Add RecallNest as an MCP server in your AI client, then call its tools such as store_memory, search_memory and resume_context from the client.
What it does
These come from the README; pickfit didn’t check them one by one.
- Share memory across all your terminalsNot tested
- Recall context saved in other windowsNot tested
- Save session checkpoints and resume contextNot tested
How far pickfit got
Linux x86_64
- Installs
- Connects
- A tool call works
- Works offline
macOS: the project says it runs there; pickfit hasn’t tried it.
Good for
- store durable memory for future windows
- recall memories at task start
- resume context in a fresh window
- import conversations from Claude Code, ChatGPT, Slack
- consolidate memories offline via clustering and pruning
How pickfit tested it
Each run starts from a clean machine and follows the steps above. How pickfit verifies
Linux x86_64
An MCP client connects and lists its tools
- Replayed from scratch
- Works offline
- Checked on 31 Oct 2026
- Version 4.0.0
An MCP client handshake over stdio listed 30 tools.