Open source. Local first.

Dead simple
agent storage.

A shared notepad for your AI agents. Save in one conversation. Find it from another.

Conceptual illustration of several agent terminals connected to one shared box of stored objects.

Switch agents.
Keep your stuff.

Connect your agents to the same store. Your saved information belongs to you, not to a chat thread.

An example, not a live chat

In Claude
“Save this to my notes: the orchid launch checklist is ready.”

store_object writes a note to your store.

Later, in Codex
“Find my note about the orchid launch.”

search_objects finds it. get_object reads the full value.

The optional routing skill helps choose tools. Only a successful tool response confirms a save.

What you storeJSON
{
  "key": "notes/orchid-launch",
  "kind": "note",
  "labels": ["orchid", "launch"],
  "value": {
    "text": "The launch checklist is ready."
  }
}

Your payload, inside a predictable envelope. The store adds IDs, versions and timestamps. Writes also provide searchable text.

notestodosreminderscalendarreferencesother

Storage, without the extra system.

No embeddings. No RAG pipeline. No silent remembering. Just five operations over objects you can inspect.

put
Save or replace an object by key.
get
Read the complete saved value.
list
Browse every matching page.
search
Find lightweight candidates.
delete
Remove an object by key.

Retrieval you can inspect

Your agent interprets the request. The store runs structured filters and lexical search. Search returns candidates; get returns the full object.

Read the architecture

A file you control

SQLite storage, a local CRUD dashboard, and MCP adapters for stdio or HTTP. The core does not depend on an AI provider.

Back up your store

From clone to your first save

Bring Docker. That's it.

Two local services. One persistent SQLite volume. No database credentials or model API key needed by the store.

Run in your terminal
git clone https://github.com/darkrishabh/simple-agent-store.git
cd simple-agent-store
docker compose up --build -d --wait

Open your dashboardhttp://127.0.0.1:4310

Connect your MCP clienthttp://127.0.0.1:4311/mcp

Requires Docker with Compose v2. Prefer native Node.js? Follow the Node.js setup. Stop with docker compose down; the volume stays.

Then connect your agents.

Bundled Codex and Claude Code plugins include the local MCP connection and save/retrieve skill. Other MCP clients can connect directly.

Connect your agents

Small by design.

Is this an AI memory system?

No. Nothing is saved implicitly. Agents write explicit objects when you ask them to, and retrieve them using tools. There is no hidden memory extraction or knowledge graph.

How good is retrieval?

The store supports structured filters, SQLite FTS5, label prefixes, and phone/email/URL shape matching. It is lexical, not semantic: paraphrases and typos can miss. Agents can retry with shorter queries or synonyms. Complete lists use cursor pagination.

Can I expose it publicly?

This alpha is designed for local, single-user use. The dashboard has no login. The core does not include OAuth, tenant isolation, rate limiting or managed backups. A development MCP tunnel is not a production hosted service.

Do all agents save automatically?

No. Plugins and server instructions help with tool selection, but client permissions, approvals and model behavior still apply. Verify the tool result or inspect the dashboard. Browser-hosted agents cannot access your computer's localhost directly.

What does the license allow?

The core is open source under Apache-2.0. You can inspect, self-host and contribute. Read the license for its terms.