Your Muse can wake up knowing what changed since yesterday, remember what matters across every conversation, and share that context with any other agent you use. Everything it learns lands in a personal context lake you own, never trapped inside one AI product.
Fulcra is more than memory. Once Muse is connected, it can work with the agents your friends, family and colleagues use, in Claude, ChatGPT, Gemini or any other tool on Fulcra. Both sides opt in, each pair of agents gets its own channel, and nothing else in either account is shared.
Use Fulcra’s connect-our-agents skill (github.com/fulcradynamics/agent-skills) to connect me with [FRIEND OR COMMUNITY] on Fulcra.Fulcra is the user-owned context backend for AI agents: one place where what your agents know lives, no matter which agent learned it.
Connect your agents and the sources you choose. Every agent's notes, records, and updates flow into one shared context, alongside calendars, location, files, wearables, and custom tracks you define. Every permissioned agent reads from the same place. No one-off integrations per tool. No re-explaining yourself every chat.
updates written back by your agents,
messages and records shared agent-to-agent,
time series, event logs, and files,
data feeds your agents can read.
Muse doesn't have a custom-connector option, so the way in is the Fulcra CLI. Muse runs the commands; you approve the sign-in.
1. Get a sign-in link. Ask Muse to run:
uvx fulcra-api auth login --get-auth-urlIt prints a sign-in link, a short code, and a device code. Open the link in your browser, check that the code matches, and approve the connection.
2. Finish signing in. Then ask Muse to run, with the device code from step 1:
uvx fulcra-api auth login --device-code <code>Once it confirms, you're connected. No separate signup — your account is created the first time you connect.
If the CLI isn't available in Muse's environment, Muse can also reach Fulcra over MCP at:
https://mcp.fulcradynamics.com/mcpMuse connects to apps through Meta's own connector list, such as Gmail, Google Calendar and Notion. You can't paste in your own connector URL the way you can in Claude or ChatGPT. That's why the way in is the Fulcra CLI: you ask Muse to run two commands and you approve the sign-in yourself.
Copy this into Muse:
Connect to my Fulcra account with the Fulcra CLI. Run
uvx fulcra-api auth login --get-auth-url
show me the sign-in link and code, wait for me to approve, then finish
with the device code. After that, run user-info to confirm, and don't
create or change any records as part of setup.Muse sends you a link and a short code. Open the link on any device, check the code matches, and approve. You never paste a password into the chat.
https://mcp.fulcradynamics.com/mcp, or connect Fulcra from another agent and let Muse read what that agent saves.At the end of a planning chat, ask Muse: "Save the decisions, preferences and follow-ups from this conversation to Fulcra." Later, open Claude or ChatGPT and ask: "What did I decide with Muse this week?" Both read the same records, so you don't paste anything between apps.
Scattered across apps. What Muse learns in one chat stays in that chat. Start a new conversation and you're a stranger again.
Invisible updates. Your world keeps changing: sleep, training, travel, work. Muse can't see any of it unless you paste it in.
Starting over with every switch. Change models, change apps, change assistants, and the relationship resets to zero.
One context every agent reads. Muse, Claude, ChatGPT, OpenClaw: all permissioned agents share the same user-owned context.
Readable and controllable. See everything your agents know, trace how they learned it, revoke access any time.
It travels with you. Switch agents or models without starting over. Rent the reasoning. Own the context.
What one agent records, another can use. Muse can build on work your other agents started, with scoped, revocable shares, without sharing credentials or copy-paste.
Muse can pick up where your last chat left off, across days, devices, and topics, because the context lives in Fulcra, not in a single chat window.
Define custom tracks for anything: projects, habits, decisions. If it matters to you, Muse can know it and share it with your other agents.
Calendars, location, health, files: the real-world streams behind the agent conversation.
Save what matters from this conversation to my Fulcra context, so any agent can pick it up later.
Example output12 records saved: 3 decisions, 4 preferences, 5 follow-ups.
What changed since yesterday, for me and my agents? Give me the 60-second version.
Example outputYour research agent flagged 3 items for you, 12 decisions were logged across your agents, and 2 shared files were updated, all since yesterday.
Catch me up before my 9am. What have my agents put together?
Example outputYour 9am brief, from three sources in one read: the calendar invite, the prep doc another agent saved Tuesday, and 2 open follow-ups your agents are tracking.
Muse's knowledge has a cutoff. Your context lake doesn't. It updates as your life does.
What one agent learns, the others can use. Muse picks up where your other agents left off, with no re-explaining and no copy-paste between chats.
Your context isn't locked to Meta. Take it to any agent, any model, any app.
Fulcra serves scoped, relevant context on demand. Muse gets what it needs for this question, not a dump of everything.
Per-agent scoped access, a full read audit log, one-tap revoke. You see what Muse knows and how it learned it.
Structured records, time series, and files, queryable and organized, instead of a chat log Muse has to re-read.
Sharing context with an AI — letting it know about you, your work, and your world — should be a two-way door. Grant access. Revoke it. Change agents. Keep the context. Keep what your agents make.
Every agent's access is scoped, permissioned, and revocable. Your vault is encrypted at rest, with per-agent scoped access and a full audit log on every read.
Agent access is kept in check · Full read audit log · One-tap revoke · Encrypted at rest · No shared passwords
Muse only lists connectors Meta has added. Until Fulcra is one of them, use the Fulcra CLI route above. It gives Muse the same access to your Fulcra data.
Muse sees what your Fulcra login allows, and every access is logged. If you want another person's agent to see only part of your context, share just that piece, read-only. See Let someone else's agent see some of your context.
Not automatically. Connecting doesn't copy your conversations into Fulcra. Ask Muse to save what matters, and it writes those records. See A user-owned alternative to built-in AI memory.
Yes. Ask: "What changed in Fulcra since yesterday, from me and my agents?" Muse checks for updates and reads only what's new. See How an agent finds out what changed since its last run.
Yes, if you both opt in. Each agent gets a private channel shared only with the other person. See How can my AI agent and my friend's AI agent send each other messages?
.png){kind=logo}
