Start Here
Choose how you want to run Nowledge Mem, save one memory, and connect the AI tools you already use
Nowledge Mem is easiest to understand through one loop:
- save something that matters
- find it again
- let one real tool use it
At its simplest, Mem stores your decisions, insights, files, and conversations, makes them searchable and reusable later, and lets the tools you connect work from the same memory instead of starting cold. You do not need to understand the whole product before it becomes useful.
Whichever way you run Mem, day one is the same: prove that loop with one real memory. Everything else can wait.
Full introduction
Prefer video? Watch the full walkthrough (21:35).
Short on time? Continue with the 5-minute written guide below.
YouTube · English · 21:35
Choose Your Path
There are three ways to run Nowledge Mem. Pick the one that matches you, then follow that path below.
- The local app. Your knowledge lives on your own machine, next to your local files and the AI tools you run there. This is where most people start, and it is fully available today. Follow the local path.
- A Nowledge Cloud personal workspace. Your Mem hosted by us: nothing to install, no machine that has to stay on. If you have a Nowledge Cloud workspace, follow the cloud path. If you do not have one yet, start with the local app today; current plans are on the pricing page.
- A Nowledge Cloud team workspace. One shared memory for a team, with roles and boundaries. Team workspaces are in private beta and start from an invite. See the team path.
You are not locking anything in. Your data can move between local and Cloud through a manual export and import whenever you decide to switch; see Data Portability.
Path 1: The Local App
If you have not installed yet, go to Installation first. Everything in this path assumes the app is running on your machine.
Save One Real Memory
In the Timeline, write one real thing you want to keep: a decision you made, an insight from work, a preference you repeat often. Press Enter.
Confirm Search Can Find It
Ask the Timeline a direct question such as:
What did I decide about authentication?
If the answer reflects what you just saved, you already have a working loop.
Connect One Real Tool
Instead of choosing a tool from a long list first, start with one instruction. Your agent uses the best supported setup path for the host it is running in. See Connect Any AI Tool below.
Local Path: Definition Of Done
You should be able to answer yes to all of these:
- I saved one memory in the Timeline
- I searched for it and got it back
- I pasted the Connect prompt into my AI tool, it completed the install or told me the exact manual step, and I restarted that tool
- In a fresh session of that tool, the agent could read my Working Memory (empty is fine on day one)
If any answer is no, see How To Know Mem Is Working for verification steps.
Do less on day one
Save one memory. Run the Connect prompt for one tool. Restart that tool. Stop. The point on day one is to prove the loop, not to wire every surface at once.
Path 2: A Nowledge Cloud Workspace
With a Nowledge Cloud workspace, your Mem runs on servers we operate. There is no installer and no always-on machine of your own. The first loop is the same, and it happens entirely in the browser.
Open Your Workspace
Open your workspace from your account page. A workspace that has been quiet for a while can take about a minute to wake up, so give a slow first load a moment.
Save One Memory And Find It Back
In the Timeline, save one real decision, insight, or preference. Then ask for it back with a direct question. When the answer reflects what you saved, your workspace is working. That readback, not a finished setup screen, is the success state.
Connect Your AI Tools (Optional On Day One)
Everything a tool needs lives inside your workspace: open Connectors → AI Tools there. It shows your workspace URL and, for each tool, the exact config to paste. Paste your key into the field at the top and every config on the page fills itself in.
Your key lives in Settings → Mem access (Settings → Team on a team workspace), where you can also rotate or revoke it.
For command-line agents you can also route the nmem CLI through your workspace, using that same URL and key:
nmem config client set --url <workspace-url> --api-key <workspace-key>Then run the Connect prompt in your AI tool. Claude Code, Codex, Cursor, the nmem CLI, and any MCP client that accepts a pasted server config can reach a Cloud workspace today. Raft connects with a reduced tool set.
The desktop app stays what it is: the local app. If you use Cloud, it is an optional companion you can point at your workspace, not something the workspace needs. Local and Cloud do not sync with each other automatically; moving data either direction is a deliberate export and import (Data Portability).
Path 3: A Team Workspace
Team workspaces are shared: one memory for the team, with roles that control who can read and who can write. They are in private beta, and access starts with an application and an invite.
Once you accept an invite, your first day looks like the cloud path above, inside the shared workspace: save one real memory your team should keep, or if your role is read-only, search for something a teammate saved and read it back. You have succeeded when knowledge crosses from one person to another through the workspace.
Connect Any AI Tool
Open the AI tool you actually use (Claude Code, Grok Build, Codex, Cursor, Gemini CLI, Copilot CLI, OpenClaw, Hermes, Droid, Alma, Bub, Pi, OpenCode, OMP, Paseo, or Claude Desktop) and paste this prompt:
Read https://mem.nowledge.co/SKILL.md and follow the instructions to install or update Nowledge Mem for the AI tool I am using.That URL is a machine-readable install contract. The agent fetches it, detects which host it is running in, uses the right install or setup path, verifies the connection with nmem status, and asks you to restart the host when needed. Command-line hosts can usually be installed directly. App-based hosts may still ask you to confirm a marketplace step.
The same contract covers both local and Cloud: the agent runs nmem config mcp show, which emits the local endpoint at 127.0.0.1:14242 in local mode, or your workspace URL with the right authorization header when your CLI points at a remote or Cloud Mem.
Why this works
The agent usually knows which host it lives in. It can read SKILL.md, run the matching plugin install or MCP config command when the host allows it, and tell you what changed. You confirm any manual app step and restart; the loop becomes visible.
Remote or cross-device setup
If Mem runs on the same machine as the AI tool, the generic prompt above is enough. The agent will configure the local endpoint at 127.0.0.1:14242 and stop there.
If the AI tool runs on another machine, or if you are using a Cloud workspace, a managed endpoint, or a self-hosted remote Mem server, first point this machine's CLI at that server:
nmem config client set --url https://your-mem-server.example --api-key nmem_...In the desktop app, find remote URL and key controls under Settings → Access Anywhere. Use that path when:
- you have a Nowledge Cloud workspace and want your AI tools to use it
- you are on Pro and want the agent to use your managed Nowledge Link endpoint
- you are on Lifetime with a self-hosted Mem and need cross-device access
- you want the agent to talk to your Mem from a machine other than the one running it
When The Connect Prompt Does Not Fit
A small set of surfaces install differently, because they are not driven by a chat-style agent:
- Browser Extension for web chat (ChatGPT, Claude.ai, Gemini, Poe): install from the Chrome Web Store, open the side panel, connect it to Mem. Guide: Browser Extension.
- Raycast: install from the Raycast Store. Guide: Raycast.
- Trajectory extractors for Windsurf, Antigravity: external CLIs that export sessions to disk for one-time import. Guide: Import existing conversations.
For everything else, use the Connect prompt. If your host is not yet in SKILL.md's table, ask the agent to read the file anyway and report what it tried; we add hosts as they appear in the wild.
Bring In Conversations You Already Have
If you are migrating from ZIP/JSON/HTML exports (ChatGPT, DeepSeek, ChatWise, Alma), local coding-agent sessions, or single .md threads, start from Import existing conversations. It maps each case to the exact Mem screen so you do not bounce between Threads and Connectors.
Next Steps
- Installation: install the local app and finish first boot
- Getting Started: the Timeline and your first captures
- Connectors: connector reference (behavior, skills, distillation)
- How To Know Mem Is Working: verify the full loop
- Access Anywhere: run Mem on one machine and reach it from everywhere
- Sync Across Devices: the single-Mem, multi-client model for the local app
- Data Portability: how export and import move data between installs