What does this automation do?
Each completed item is checked against your definition of done. Genuinely finished work is closed out; the rest is reopened with a follow-up.
What can you do with it?
- Catch work that was ticked off but not finished
- Reopen it in Google Tasks rather than filing a new item
- Close out the work that really is complete
The Steps
This automation runs in 3 steps:
- When Google Tasks — Task Completed
- Then Taskade AI — Ask an agent ("Definition of Done")
- Then Branch — Not actually finished: Re-open Task, Add Task · Genuinely done: Mark Task As Completed
What makes this different from a Zapier-style flow?
The "Definition of Done" step is not a one-off model call. It is a Taskade AI agent that knows your acceptance criteria and what counts as finished, remembers the items it has reopened before and why, so every run builds on the last one instead of starting from an empty prompt.
A stateless "ask the model" step forgets everything the moment it returns. An agent keeps its knowledge and its memory between runs, carries its own tools, and can be handed work by other automations.
Who is this automation for?
Built for teams where "done" and "done done" are different things. Each completed item is checked against your definition of done. Genuinely finished work is closed out; the rest is reopened with a follow-up.
How do you set it up?
- Open the "Automations" tab in your workspace.
- Click ➕ Add automation, then start from scratch.
- Add the trigger "Task Completed" from Google Tasks.
- Add the action "Ask an agent" from Taskade AI. Point it at an agent called "Definition of Done".
- Add a Branch step: Branch — Not actually finished: Re-open Task, Add Task · Genuinely done: Mark Task As Completed.
- Enable the toggle in the top-right corner to activate the automation.
Prefer to describe it instead? Click Build this flow with AI above and Taskade wires the trigger, the steps, and the agent for you.