For AI agents: The documentation index is at https://docs.digitalocean.com/llms.txt. Markdown versions of pages use the same URL with index.html.md in place of the HTML page (for example, append index.html.md to the directory path instead of opening the HTML document).

Celiums Memory turns any DigitalOcean droplet into a self-hosted memory layer for AI agents — exposing

semantic recall, an append-only signed agent journal, and a 5,100-module expert knowledge base over the

Model Context Protocol (MCP).

It connects in one click from Claude Web/Desktop, Claude Code, Cursor, Antigravity, Windsurf, ChatGPT

and any MCP-compatible client; tools like recall, remember, forage, journal_*, research_* and

write_* become available to your model the moment the droplet boots.

The stack is fully local: PostgreSQL 17 + pgvector, Qdrant for vector search, Valkey for

caching, an embedded REST + MCP engine, and a Crystal-UI dashboard with BYOK for the LLM provider of

your choice. Your memories, embeddings and journal entries never leave the droplet you control.

Provision the droplet, SSH in once to run sudo celiums-setup to expose the dashboard via a domain (Let’s

Encrypt) or directly by IP (self-signed) — and your private memory server is live.

Software Included

Creating an App using the Control Panel

Click the Deploy to DigitalOcean button to create a Droplet based on this 1-Click App. If you aren’t logged in, this link will prompt you to log in with your DigitalOcean account.

Creating an App using the API

In addition to creating a Droplet from the Celiums Memory AI 1-Click App using the control panel, you can also use the DigitalOcean API. As an example, to create a 4GB Celiums Memory AI Droplet in the SFO2 region, you can use the following curl command. You need to either save your API access token to an environment variable or substitute it in the command below.

curl -X POST -H 'Content-Type: application/json' \
         -H 'Authorization: Bearer '$TOKEN'' -d \
        '{"name":"choose_a_name","region":"sfo2","size":"s-2vcpu-4gb","image":"celiumssolutions-celiumsmemoryai"}' \
        "https://api.digitalocean.com/v2/droplets"

Getting Started After Deploying Celiums Memory AI

1. SSH into your droplet

ssh root@your\_droplet\_public\_ipv4

The MOTD will guide you to the next step.

2) Run the one-time setup wizard

bash  

sudo celiums-setup  

Choose how the dashboard is exposed:

• FQDN mode — point a domain (e.g. memory.example.com) at your_droplet_public_ipv4, enter it in

  the wizard, and Caddy fetches a free Let’s Encrypt certificate automatically.

• IP mode — uses https://your_droplet_public_ipv4 directly with a self-signed certificate. Your

  browser will show a one-time warning the first time you visit.

  When the wizard finishes it prints your dashboard URL and your engine API key (also saved at

  /root/.celiums/api-key).

  3. Create your admin user

  ---

  Open the URL the wizard printed (or https://your_droplet_public_ipv4), set a username + password, and

  copy the API key from the Settings page.

  4. Connect any MCP client

  ---

  Paste the configuration snippet into your editor or assistant — works with Claude Web, Claude Code,

  Cursor, Antigravity, Windsurf, ChatGPT and any MCP-compatible host. Path table and ready-to-paste JSON for

  every host:

  https://github.com/terrizoaguimor/celiums-memory/blob/main/docs/integrations/mcp-clients.md

  Reference

• Logs: journalctl -u celiums-engine -u celiums-dashboard -u celiums-proxy -f

• Config: /etc/celiums/env (mode 0600, owner celiums)

• Repo: https://github.com/terrizoaguimor/celiums-memory

• Issues: https://github.com/terrizoaguimor/celiums-memory/issues