Agentic AI Users
An AI Agent can also work and function as a regular user. Give it a user account of its own and it becomes a worker in its own right — it takes tasks, records its time against them as it works, and everything it does is kept under its own name. Work written over MCP is stamped as agent work in the record, so the ledger always shows which entries came from an Agent and which came from a person.
When to use an Agent User
When an Agent works alongside you — in your editor or a chat window — connect it as yourself, and its time is your time (see Connecting an agent). An Agent User is for work that isn't yours: an autonomous Agent that picks up tasks and works them on its own. The account keeps that work under the Agent's own identity, separate from any person's record.
Setting one up
Create the account — invite the Agent under Personnel, the same way you'd invite a person. An ordinary member account is all it needs.
Mint its token — sign in as the Agent and create its personal access token under Edit Profile > API Access. Agent access must be enabled for the organization, the same switch as always.
Register the connection — configure the Agent's MCP connection with that token, exactly as described on Connecting an agent. From then on, everything it does over MCP is recorded as the Agent.
How its work is recorded
The Agent works the same surface people do. It takes a task, records minutes against it as the work happens, and its entries arrive as drafts that become recorded time through acceptance. Accepted agent time sits in the same ledger as everyone else's — reports, billing, and audits see the task, the minutes, and the Agent's identity behind them.
For instance, if you use Claude Code/Cursor or any other interface that supports MCP, you can instruct the AI Agent via prompts to use it's MCP account to perform any/all of the following:
record any work it performs during your session into time entries and/or tasks
plan out and organize a project for you
pick up and complete any tasks it has been assigned
"watch" a particular project or stage for any tasks that appear, then complete and close them
Handing work back to people
When the Agent is blocked, or has findings a person needs to look at, it can park the task: marked blocked with a note saying what it's waiting on, and assigned to the person who needs to act. The board shows exactly where agent work is waiting on a human.
A person's drafts stay private until accepted — that never changes. An Agent User's record carries no such expectation: its work exists to be reviewed by the people it works with.
The tools an Agent User works with are the same ones on Tools. Endpoint shapes, status codes, and request fields are in the API Reference.