All your coding agents.
One personal AI.
Jaz runs the coding agents you already use, on the plans you already pay for, and lets them work together, with memory, Tasks, a CRM and your accounts around them.
Ask anything…
Agent for this thread
Every agent, as itself.
Jaz runs coding agents over the Agent Client Protocol, so you get the agent its lab built, with the subscription you already have. Seven work out of the box, and any other ACP agent plugs in.
- Your plans
- Sign in with ChatGPT, Claude, Grok, Kimi, Google or Meta, or paste an API key.
- Native behaviour
- Each agent keeps its own prompts, tools, compaction and context limits.
- Per thread
- Pick the agent for each thread and switch when a better one ships.
Everything around the agents.
A coding agent is a great worker with nowhere to live. Jaz gives it one: a place to remember, people and work to track, accounts to act with, and routines that run while you are away.
Agents that work together.
A bot is an agent with a name, a face and its own home folder, running whichever agent and model you choose. Put Claude Code, Codex and Grok bots in one group, @mention who should answer, and let them build on each other's work. Routines run them on a schedule, or the moment an email, WhatsApp, Telegram or Slack message arrives.
Long jobs go to background workers, and each bot shows up separately in Usage.
Today 9:41
Inbox created a routine · Triage new Gmail
Scout is reading the changelog…
Morning briefing
Every day at 7:30 · next tomorrow 7:30
Triage new Gmail
Running now · When a Gmail message arrives
Reply to WhatsApp
When a WhatsApp message arrives
Weekly spend report
Fridays at 17:00 · next Fri 17:00
Deploy check
On webhook · last run 12m ago
Tasks and a CRM, built for agents.
Two open-source apps that connect over MCP and open as tabs inside Jaz. Your agents file issues, update deals and draft follow-ups in the same place you look at them.
Tasks
tasks.jaz.chatAn issue tracker that feels like Linear: issues, projects, board and timeline views, and the keyboard shortcuts you already know.
- Linear-compatible API
- MCP server
- Self-hostable
- MIT
CRM
crm.jaz.chatA CRM that builds itself from Gmail, Calendar and meeting transcripts, with pipelines, notes in nested pages, and drafted follow-ups for the people you owe a reply.
- Pipelines
- Notes and pages
- Follow-up drafts
- Self-hostable
- MIT
Memory that survives the thread.
Jaz remembers what matters: who's who, what you're working on, how you like things done. Agents pick it up as they work, and Jaz dreams overnight, while you sleep. Every fact keeps a source you can check.
It's all plain files on your machine. Open them, search them, back them up.
Your accounts, in reach.
Connect your accounts once and every agent and bot can search and act with them. Any MCP server adds its tools for every agent, and MCP Apps add their own tabs. Skills are markdown playbooks on disk: type $ and they're one keystroke away.
Sessions and tokens stay on your machine. Several accounts per service.
Connections
Connect accounts for agent tools and memory.
Connected
Gmailsam@lumen.dev · Synced 5m ago
Google Calendarsam@lumen.dev · Synced 5m ago
SlackLumen workspace · Synced 2m ago
WhatsAppLinked device · Synced just now
Telegram@sam · Synced 1m agoProductivity
Google DriveSearch and read your Docs and filesMCP servers
Remote Streamable HTTP connections available to Jaz and capable ACP agents.
https://tasks.jaz.chat/mcp
https://crm.jaz.chat/mcp
https://mcp.context7.com/mcp
https://mcp.sentry.dev/mcp
https://mcp.linear.app/mcp
Skills
A browser beside every thread.
Agents open pages, click and fill forms while you watch their cursor, in the side browser, your own signed-in Chrome, or a Chromium running in the background.
Two quiet places have a table for four at 20:00. Casa Azul has the garden, so I'm booking that one.
Claude is choosing 20:00 at Casa Azul…
A client for you. A server for the work.
Jaz splits in two. One half is the app you open — on any of your computers — to talk, review, and steer. The other keeps running on your Mac, a home server, or a small VM, taking in integrations and running routines with no window open.
On your laptop — review, steer, and talk to agents.
Your devices
The window you use
Open Jaz from any paired computer — the same live workspace every time.
- Talk to agents and review threads
- Steer bots, boards, and memory
- Pick up on any device later
Server
The worker that stays on
Leave it awake and the background work keeps going, on machines you control.
- Runs bot routines on schedule
- Listens to your integrations
- Writes files, memory, and audit trails
Always-on mode
Run it where it can stay awake.
Local when you're hands-on; on an always-on machine for 24/7 work. Your devices come and go — the work keeps its place.
Integrations keep arriving
Gmail, Calendar, Slack, WhatsApp, Telegram, and MCP servers keep delivering work.
Routines keep their schedule
Briefings, triage, and reminders run with no window open.
Memory keeps up
Results land in your files, so the next device opens to the current state.
Own your personal AI.
You should own your personal AI: the system, the memory, the data. Models come and go — Claude today, whatever is best tomorrow — but everything they learn about you stays in plain files on machines you own. This is too profound a technology to be locked inside someone else's tool. So Jaz is open source — yours to read, audit, run, and change.
Your machines. Your files. Any model. No lock-in.
If Jaz resonates, star it on GitHub.