AI agent that scopes a feature request into an estimated backlog item
Turn feature requests from a Fluent Forms form into ClickUp tasks written by an AI Agent, with a title, acceptance criteria, a rough effort estimate and a size tag, then tell the team in Slack.
Fluent Forms
AI Agent
ClickUp
Slack5 steps Setup ~8 min Used on 3.2k sites Productivity Free to import
What this workflow does
Feature requests tend to arrive as one long paragraph. Before anyone can plan the work, someone has to give it a title, work out what done looks like and guess how big it is. This workflow hands that first pass to an AI Agent, so a request reaches ClickUp already shaped like a backlog item.
A visitor sends the request through a Fluent Forms form. FlowMattic passes the sender's name and message to the AI Agent. Its instructions ask for a clear title, acceptance criteria as a checklist and a rough effort estimate, and tell it to create the item in ClickUp. The agent has two ClickUp tools: Create Task, where it writes the task name and description, and Add Tag to Task, which it is meant to use for the estimated size.
Once the agent finishes, a Slack step posts a short note that a new scoped item came in, naming who sent it. The estimate is the model's guess from the request text. Treat it as a starting point for your planning meeting, not a commitment.
Public feature request page
Customers fill in a Fluent Forms form on your site. Each request becomes a ClickUp task with acceptance criteria, so product managers start from a draft instead of a raw message.
Internal ideas form for staff
Support and sales staff submit ideas they hear from customers. The agent writes each one up the same way, which makes the backlog easier to compare and sort.
Client change requests for an agency
Clients ask for changes through a form. The task arrives in ClickUp with a size tag, which helps you decide what to quote or schedule.
How this workflow runs
What happens at each step once the workflow is on, and what you need before you import it.
Fluent Forms New Form SubmissionStarts a run for each feature request sent through your Fluent Forms form. The name and message fields are sent to the agent, and the name is used again in the Slack note.
- AI Agent Execute Agent
The AI Agent reads the request and, following its instructions, writes a title, a checklist of acceptance criteria and a rough effort estimate, then uses its ClickUp tools. The template uses OpenAI GPT-4o; connect your provider on import.
- ClickUp Create Task
The agent's tool for creating the backlog item. It fills in the task name and description itself. Connect ClickUp and run a test to confirm where the task is created.
- ClickUp Add Tag to Task
The agent's tool for tagging the new task with its estimated size. The agent writes the tag text. If your team uses fixed size tags, list them in the agent's instructions.
- Slack Send Channel Message
Posts New scoped backlog item from, followed by the sender's name, to Slack. Connect your workspace and send a test to confirm the message reaches the channel you want.
What you need
- FlowMatticCore
- Fluent Forms PluginPlugin
- AI ProviderAccount
- ClickUp AccountAccount
- Slack AccountAccount
After importing, map your own data. To keep the template universal, not every step is pre-filled. Open each step in the builder and connect your accounts and fields (channels, lists, sheets, recipients) before you switch the workflow on.
Make it yours
Changes worth making after you import it.
- Name your size scale in the agent's instructions, for example S, M, L and XL, so the tags stay consistent and you can filter ClickUp by them.
- Add the agent's response to the Slack message. Map it in the message text so the team can read the scoped item without opening ClickUp.
- Put a Filter after the trigger if the same form takes other kinds of messages, so only real feature requests reach the agent and ClickUp.
Questions about this workflow
Which model writes the backlog items?
The template is set to OpenAI GPT-4o. You connect your own AI provider account on import and can choose a different model in the agent step.
Will every request become a ClickUp task?
That is what the agent is told to do. Because the agent decides when to call its tools, check Task History during testing to confirm a task is created for each run.
How accurate is the effort estimate?
It is a rough guess based only on the text in the request. Use it to sort the backlog, then let the people doing the work set the real estimate.
Does my form need special fields?
The template maps a name field and a message field. If your form uses other names, remap them in the agent's input and in the Slack message.
More Project & Task Automation workflows
Other ready-to-import templates from the same category.
Import this workflow into your WordPress site
Download the JSON, import it from the Workflows screen and connect your accounts. Every run stays on your server, with no per-task fees.
30-day money-back guarantee Unlimited workflows and runs Self-hosted