marvis
EN IT
★ GitHub Get started
The Company Brain for AI agents

Your agent remembers facts. The brain tracks the work.

Built-in memory and CLAUDE.md are just notes — they reload whole, go stale, and don't track what you're working on.

Or: see the Console in action ↓

Local-first · Free, open-core BSL 1.1 → Apache 2.0 · Claude Code · Codex · Cursor, via MCP

The cockpit · free, local

Stop guessing what your agents did. Watch it.

Open the desktop app and your brain is right there — local, single-user, yours. Scroll to walk the four surfaces an operator lives in.

Marvis Console
scroll to explore
  1. 01 · Journal

    Every morning, last night's work — narrated.

    The brain digests everything your agents shipped and writes it up: what moved, what got decided, what's waiting on you. You read three sentences, not a 400-line log.

    brain journal · written on your schedule Marvis Console — Journal
  2. 02 · To-dos

    Then, what needs you.

    The brain bounces the calls only you can make to your queue — a price that doesn't match, a rule that changed: decide, approve, review, or delegate it back to an agent.

    your queue · decisions chained to audit Marvis Console — To-dos
  3. 03 · Tasks

    A board you share with your agents.

    To do → In progress → To approve → Done. Agents own the work and push it back when it needs a yes. Every card carries its project, its ICE-D priority, and the brain's warnings.

    kanban · agent + you, one queue Marvis Console — Tasks
  4. 04 · Project

    One initiative, everything linked.

    Open a project and the brain hands you its whole context: the journal, the decisions (and which superseded the last), what it knows, the docs, the people — not lost in Slack threads.

    project brain · decisions · docs · people Marvis Console — Project
  5. 05 · Universe

    Your whole stack as one graph.

    Projects, tasks, documents, decisions — linked across every repo. Ask "what breaks if I change this?" and get an answer that spans the company, not just the file in front of you.

    knowledge graph · cross-project X-ray Marvis Console — Universe knowledge graph
The problem

It's not memory you're missing. It's the work.

Your agent can recall facts. What it can't see is the state of things — what's in progress, what's blocked, what depends on what. So it reloads notes inside the session, and leaves a pile of disconnected docs around it.

01 · Inside the agent

Every session reloads. Nothing carries.

The whole file dumps back in, stale and entire. The agent quietly contradicts a past decision or trips a forgotten guardrail — and none of it traces back to you.

02 · Around the agent

Your notes are a bucket, not a brain.

Notion, Obsidian, a CLAUDE.md per repo — every tool stores whatever you dump in. None track what's in progress, or what breaks if it changes.

How it works

Give the link to your agent. It installs itself.

Your agent runs the wizard — no dashboard, no manual migration. Marvis lives inside the stack you already use.

01
You · 5 seconds

Give the install link to your agent.

From inside Claude Code, Codex, or Cursor — the same terminal you already work in.

$ claude run "install marvis from justaskmarvis.com"
02
Agent · runs the wizard

Your agent fetches the harness and wires itself up.

Folder scaffold, governance hooks, MCP server. The agent updates its own CLAUDE.md / AGENTS.md so future sessions wake up brain-aware.

→ cloning marvis-harness …
→ updating CLAUDE.md, AGENTS.md
→ installing 5 governance hooks
→ wiring MCP server at .marvis/mcp.sock
✓ harness in place
03
Marvis · migrates your work

In-progress projects are ingested into the brain.

Existing repos, decisions and transcripts — indexed into the graph, your constitution applied across every project at once.

→ scanning ~/code · found 12 projects
→ ingesting · 1,847 nodes · 4,212 edges
→ constitution applied to 12/12
✓ brain online
04
You · keep working

Same terminal. Now with memory, governance, audit.

Every session reads the brain; every result writes back, tagged with your approval.

$ claude
› Brain: 3 unresolved decisions from last week.
› Brain: onboarding agent broke a rule — flagged.
› Brain: new learning extracted from 4 converged sessions.
› _
Pricing

Free on your machine. Paid to have the brain on the cloud.

Self-host the whole thing for free. You pay when you want the brain always on in the cloud, or when a company needs departments and governance.

★ Free forever
Open-core · self-hosted

Marvis

Free

Your machine. Single-user.

  • Local Console + Desktop App included
  • Agent-native knowledge graph, your iron
  • Constitution + immutable audit chain
  • Connects as an MCP to Claude Code · Codex · Cursor
★ Star on GitHub → Install in 3 steps
Managed
Cloud · always-on

Marvis Cloud

from €19/mo

The brain that never sleeps. Single-user.

  • Everything in self-hosted, hosted for you
  • Always-on nightly brain cycles, laptop closed
  • Your knowledge in the cloud: use it from laptop, phone or tablet
  • Connects to Claude Code · ChatGPT · Cursor
  • EU-hosted · GDPR · audit trail
Request access →
By contract
Enterprise · self-host or private cloud

MarvisX

Custom

Multi-user. Sovereign. Whitelabel.

  • Everything in Cloud, plus:
  • Multi-user · departments · scopes
  • SSO & SCIM
  • EU AI Act Art. 12 readiness
  • Data sovereignty · whitelabel · SLA
Talk to us →
For enterprises

MarvisX — AI at full power, under control.

For companies that want AI to ship work — without giving up uptime, departmental separation, or where the data lives.

01 · Always-on workspaces

Terminals hosted on your server. Always-on, server-side.

Your team's agents run server-side, not on laptops. No "my Mac died at 3am" stories. Reachable from anywhere, audit-logged everywhere.

hosted · server-side
02 · One brain. Many teams.

Cross-department knowledge. RBAC + SSO.

Product knows what Sales decided. Engineering sees Compliance constraints. Departments share context without sharing everything. SSO, SCIM, role-based scopes.

RBAC · SSO · SCIM
03 · Your data, your model

Local storage. Local LLM, or your keys.

Data stays on your servers, never ours. Run an in-house LLM, or BYOK against OpenAI, Anthropic, Mistral. No vendor lock-in on infra or model.

local · BYOK · self-host
04 · Compliance ready

Pass AI Act audits without scrambling.

Every action chains back to the human who said yes. Logging and human-oversight requirements on by default. Your legal team writes policy; MarvisX enforces it.

EU AI Act · audit-ready
FAQ

Common questions.

Things people ask before installing.

Where does my data live?
On your hardware. Marvis is self-hosted by default. No vendor cloud, no third-party storage.
What agents work with Marvis?
Claude Code, Codex, and Cursor today. Any MCP-aware agent works out of the box. New runners take an hour to wire up.
Is Marvis open source?
It's open-core. The Marvis you install is free to fork, inspect and self-host, and it stays that way. The paid tiers sit on top — they don't close the core.
What's the difference between Marvis and MarvisX?
Same brain. MarvisX adds multi-user governance, SSO, compliance audits, and a contract — for companies that need them.
How do I uninstall?
One command rolls back the harness and exports your brain as plain Markdown. Your decisions stay yours. No hostage data.
Does Marvis need internet?
Your brain and data stay on your machine. The only things that leave are opt-out usage telemetry and a version check for updates. Your agents still need internet to reach their model providers.

Give your agent the link.
We'll meet you in your terminal.

You don't change tools. You don't migrate. You drop in one line and your stack wakes up with a brain.

★ Star on GitHub → Talk to us
How can Marvis help you? →