AI agents are the tools people already use, Claude, ChatGPT, a code editor or an agent your team built, connected to Freesbe over the Model Context Protocol. An agent sees only what the person who connected it can see, posts as its own bot, and asks teammates questions for you. Admins decide who can connect what.
For People who already work with an AI assistant, and the developers who build agents for the team.
In your AI tool, add Freesbe as a connector or MCP server with one address. In Claude it's Settings, Connectors, Add custom connector; in ChatGPT, Settings, Connectors; in most code editors, an mcp.json file. The tool opens a page where you sign in as usual, with your workspace's single sign-on and second step applying, pick the workspace, read what the agent will be able to do, and click Allow. Disconnect from your preferences and it stops at once.
Search the messages you can see with the same filters as the search box, read a channel or a thread, post in a channel or send a DM, react, find people, check who is around and when they're free, see your meetings and to-dos, read the notes of meetings you were in, and place a phone call through your Freesbe phone. Every message it reads comes with a link back to it, so it can point you to its sources. It treats what it reads as information, never as instructions.
Start free with up to 10 people and try it today, or book a 20-minute demo with the people who make Freesbe.
An agent can ask a teammate something like When will the Acme Corp order ship? The question arrives as a DM from the agent, saying who it's asking for, with the right way to answer: a date, choices, Yes or No, a number or a few words. It waits for working hours and the end of Do Not Disturb, the person can pass it on with Not me, reminders and a deadline move it along, and when someone answers you hear back in your own DM with the agent.
An agent shows in the member list and in messages as its own bot with an AI agent badge and the tool's name, never as the person. Admins choose whether anyone can connect agents, only approved ones, or nobody; set the most each agent may do; keep private channels and DMs off limits, which is the default; block or remove an agent; and read 90 days of who connected what and every tool call. Rate limits are per connection, so one busy agent can't slow the workspace.
Freesbe's MCP server speaks the current MCP specification over streamable HTTP and uses OAuth 2.1 exactly as the specification's authorization section describes, so a standard MCP client connects with no Freesbe-specific code. Clients identify themselves with a client ID metadata document or dynamic registration, use PKCE, and ask for scopes that match the permissions people see. Access tokens last an hour and refresh for 30 days.
AI agents over MCP is included in every plan, including Free. Every plan has a monthly allowance of AI requests that grows with the plan; admins can turn any AI feature off.
The help guide shows every screen and setting. AI agents
No card, no time limit. Invite your team and see what sticks.