Teams & security
For companies
TODO for AI gives a company one board where people and AI agents work the same TODOs. The agents know the business (offer, customers, voice), propose work, and do it in the company's own tools and machines.
What changes for a team
| Before | With TODO for AI |
|---|---|
| Each person chats with their own AI, results stay in private sessions | Every TODO, step and result is on the shared board |
| Prompts re-explain the company every time | One brand page and voice per project, read by every agent |
| Agents only answer; people copy results into tools | Agents work in Gmail, Drive, the CRM, the browser, servers |
| Separate subscriptions per person and per model | One plan, every model, one shared budget |
| Work happens when someone asks | Agents propose TODOs from the business context; scheduled TODOs run on their own |
Rollout
- Create the team. The owner signs in with the company email and adds the website. Get started →
- Review the brand page and voice. Settings → Context. Fix anything wrong; every agent reads it.
- Connect tools and machines. Connectors for shared inboxes, Drive, Slack, CRM. A Bridge on a server for internal systems. Machines & connectors →
- Set permissions. Defaults allow shell and file access on attached machines. Switch tools that send, pay or delete to ask. Agent permissions →
- Invite people, or turn on domain auto-join. Teams →
- Choose seats. Standard for most people, Premium for heavy users. Track spend in Account → Usage.
Projects per department
One project per team or client keeps brand, voice, files and machines separate:
Typical work
| Team | Examples |
|---|---|
| Marketing | SEO pages, social posts in the brand voice, competitor tables, ad copy, newsletters |
| Sales | Lead research, CRM updates, follow-up drafts, proposal decks |
| Support | Draft replies from the help center, tag and summarise tickets |
| Finance / ops | Collect invoices from email, reconcile exports, monthly reports |
| Engineering | Code changes on your repos and servers, deploys, log checks; also inside Zed, JetBrains and VS Code |
More examples: Productivity guide.
Before you roll out: checklist
- DPA requested from privacy@todofor.ai if you process personal data
- Sub-processor list reviewed (GDPR); restricted model set requested if needed
- Permissions changed to ask for sending, paying and deleting (defaults allow them)
- Shared accounts connected by an owner or admin, personal ones by each member
- Private work kept in My Private groups or personal projects
- Budget: seats chosen, usage reviewed after the first week
Current limits
SSO/SAML and SCIM, user-facing audit export, assignees and due dates are not available yet. See Teams and Security.
Contact
Enterprise plans, invoicing in HUF, custom provider sets: support@todofor.ai.
Last modified on