Install

Three steps, about two minutes. You install it, sign in, and connect it to your AI. Requires Python 3.10 or newer.

1. Install it

Pick either one. Both install everything you need in one go.

Option A, with pip:

$ pip install 'sibyl-memory-cli[mcp]'

Option B, one-line installer:

$ curl -fsSL https://sibyllabs.org/install | sh

Both pull in the full toolkit — sibyl-memory-client (the local SDK), sibyl-memory-hermes (the Hermes provider), and sibyl-memory-mcp (the MCP server that sibyl setup wires into Claude Code and Codex). If you see an "externally managed environment" error, jump to If something did not work at the bottom for the one-time fix.

2. Sign in

Run this. It opens a sign-in page in your browser.

$ sibyl init

Sign in there with a wallet, or with your email and a code. The terminal picks it up on its own. The free plan gives you the full memory system at no cost, no card needed. Credentials are written to ~/.sibyl-memory/credentials.json at mode 0600.

Free is the default

You do not need to pay to build real things. Free tier gives you the full five-tier memory model with a 2 MB local cap. Staker and subscription tiers lift the cap and add self-learning and the memory linter. See Tiers & access.

3. Connect it to your AI

This finds your AI app and connects Sibyl to it for you. No settings to edit.

$ sibyl setup

Then restart your AI app. That is it, Sibyl is connected.

It connects automatically to Claude Code, Codex, and Hermes. If you use a different app, sibyl setup wires the ones it recognizes; for anything else, add Sibyl in your app's own memory or connections settings. Either way your AI gains three new abilities: it can save, recall, and search what it remembers. You can also name one target or preview the changes: 

$ sibyl setup claude-code    # wire just one
$ sibyl setup --dry-run      # print what would change, write nothing

Targets: hermes, claude-code, codex. See Integrations for what each one does.

4. Test it works

Open your AI and tell it something worth remembering, like: "remember that I like short, direct answers." Close it, open it again later, and ask: "how do I like my answers?" It will know. That is the whole thing, your AI remembers now.

Bring existing memory with you

Already have agent memory in CLAUDE.md, AGENTS.md, a Codex config, or Hermes files? sibyl migrate moves it in without risk. It backs up every file it finds and byte-verifies the copy before anything else happens, wires Sibyl into each detected harness, then prints a prompt you run in your own agent to extract structured memory from the backup. The extraction runs locally; Sibyl Labs never sees your files. 

$ sibyl migrate
Backup-first, never destructive

Trimming the originals is the last step, it is optional, and it only happens after you confirm and only because a verified backup already exists. Keep the backup until you have confirmed everything migrated.

Just the SDK

If you only want the library to call from your own Python:

$ pip install sibyl-memory-client

Then jump to How it works for the API surface.

Verify

$ sibyl status     # local credentials, DB size, server-side tier
$ sibyl health     # provider self-check (schema version, DB path, tenant)

If something did not work

The steps above work for almost everyone. If the install showed an "externally managed environment" error, your system wants a sandbox first. These two lines make one and install Sibyl inside it: 

$ python3 -m venv ~/.sibyl-memory/venv && source ~/.sibyl-memory/venv/bin/activate
$ pip install 'sibyl-memory-cli[mcp]'

Then carry on from Step 2 (sibyl init).

Connected nothing in Step 3? Make sure your AI app was open, then run sibyl setup again.