Task Demon
LoginSign Up

Introducing Re-plan and Continuous Characterization

Today we are introducing two new features: Re-plan and Continuous Characterization. Re-plan allows the Task Agent to read, assess and edit plan documents made by Claude Code. It is a game-changing capability for the Task Agent, enabling a cavalcade of new workflows.

Characterization is part of the secret sauce that makes Task Demon do what it does so well. And today, Characterization becomes Continuous Characterization, automatically updating your Characterization documents as you implement tasks. This ensures that as you add new code to your codebase, your Characterization documents are always up to date, so Task Demon's Agents can continue to make the best possible decisions.

Check out the short video below to see both new features in action:

If you want to see the future of agentic coding, this video will be 10 minutes well spent

Re-plan

The Task Agent got way smarter this week, gaining the ability to re-plan a task if it feels that the first plan wasn't good enough, or if you tell it that you want to change things. The Task Agent automatically reads plan documents the moment they are submitted, analyzes them and gives you a quick summary of what the plan proposes. Task Agent gives the plan a rating and provides detailed feedback on what needs to be improved.

If it feels that a re-plan is warranted, it will do so automatically up to 3 times, though you can ask it to re-plan any time you like.

Real re-planning example

Here's an example of an actual task agent session where we re-plan a task. In this case I wanted a button that would regenerate the plan background document if it exists already, but I didn't like the first iteration of the plan so I asked it for some tweaks.

The UI below is a read-only view of the conversation between the Task Agent and the human driving it (me). The buttons are not interactive, but you can scroll through the short conversation to see re-planning in action:

Task AgentTask Chat

Triage has been completed; the task has been classified as coding

I am preparing a background document to assist the planning process, I will let you know when it is finished

Background Document Completed

Success
I've finished the background document. Next I'll start the plan workflow.
View Background Document

Plan Successfully Completed

Plan
The plan is complete. I'll assess it now and let you know what I think.
Copy the prompt to use with your coding assistant.
View Plan Document

I'm reading the plan document, just a moment...

Plan Assessed

Overall Rating:

Task Overview: Introduce a “Regenerate Background Doc” button on the Task Details page that invokes a new backend API. The API re-runs prepareTaskBackground(), triggers a real-time Pusher event, and returns a success JSON response. The UI button shows a loading spinner and disables during the request, with toast notifications on success or failure.

Implementation Steps: 1) Review the existing assess-plan endpoint. 2) Create a POST /api/tasks/[taskId]/regenerate-background-doc route with auth, service call, and Pusher trigger. 3) Write backend tests for the new route. 4) Extend TaskActionButtons.tsx with imports, state, the new button, and a handler for API calls and toast notifications. 5) Write UI tests to verify loading states and toasts. 6) Verify (and optionally test) the TaskProvider Pusher event handling.

ok I don't want it using pusher for this though, I just want the new API endpoint to block until the new document is generated - this is usually only 10-20 seconds so should be reasonable. During this time the button should be in a disabled state showing a spinner. Also, to be clear, I want it to replace the content of the existing Background Document for the Task with the newly-generated content (or create a new Background Document if somehow one does not exist on the Task, for instance if the user deleted it). When the API endpoint returns the response, the button is enabled again and the update is complete. There should be a pusher message informing that a document was updated (or created), and the UI should reload the documents in response to that (I know that already happens to at least some extent)

Plan Update Requested
I have requested an update plan. Here's the feedback I gave it:

Update the plan to remove the use of pusher for the new API endpoint's operation. The new API endpoint should block until the new background document is generated, which takes about 10-20 seconds. During this time, the UI button should be disabled and show a spinner. The API should replace the content of the existing Background Document for the Task with the newly generated content, or create a new Background Document if none exists. When the API returns, the button is enabled again and the update is complete. A separate pusher message should still inform that a document was updated or created, and the UI should reload the documents in response to that event as it currently does.

The plan has been updated to reflect your requirements:

  • The new API endpoint will block until the new background document is generated (about 10-20 seconds).
  • During this time, the "Regenerate Background Doc" button will be disabled and show a spinner.
  • The API will replace the content of the existing Background Document for the Task or create a new one if none exists.
  • When the API returns, the button will be enabled again, marking the update as complete.
  • A separate pusher message will still inform that a document was updated or created, and the UI will reload the documents accordingly.

Is there anything else you would like to adjust or add before we proceed?

New plan document has been created, just a moment while I double-check it...

Plan Assessed

Overall Rating:

The plan introduces a new synchronous POST /api/tasks/[taskId]/regenerate-background-doc endpoint that calls the existing prepareTaskBackground service, blocking the request for ~10–20 seconds while regenerating or creating the task’s background document. The service layer emits a Pusher 'task.updated' event once regeneration completes. On the frontend, the TaskActionButtons component adds a “Regenerate Background Doc” button, manages loading state with a spinner and disabled state, and shows success/error toasts. The TaskProvider listens for the Pusher event to reload and display the updated document.

Implementation is broken into detailed subtasks: preparatory environment checks, backend service update for Pusher emission, API route creation without inline Pusher, interface definition, backend and UI tests, UI component updates (button, handler, toasts), and verification of the real-time provider handling.

The story of this task

As soon as a Task is created, the Task Agent triages it (result was "coding" in this case), then creates a background document that will be fed to the planning agent to help write a really good plan

Once the Background Document is created, the planning agent creates a plan document, which the Task Agent reviews. It gave it an excellent rating and provided a high level overview of what the plan proposed

Then we throw it a curveball - after reviewing the initial plan it looked like it was not quite doing things right, so some clarifying and refining guidance was provided by us, resulting in a re-plan

Task Agent takes our updated guidance and requests a new plan document with specific feedback on what it needs to see improve

2 minutes later, the new plan document rolls in, it gets another excellent rating, and Task Agent gives a quick summary of what it now proposes

Re-plan is available in the Task Agent today for all users. Shipping as part of this update is the ability for Task Agent to read each version of the plan and answer your questions about it - try it out!

Continuous Characterization

One of the ways Task Demon stands apart is through its Characterization process - an automated multi-agentic process that automatically analyzes and characterizes your codebase. It's always recommended to run Characterization when you start a new project in Task Demon so that it knows how your project works, but what about changes you make after that?

That's where Continuous Characterization comes in. Now, every time you click the "Mark Implemented" button on a Task, the Characterization Agent will be informed. It will be passed the latest plan document for the Task and will automatically decided whether any other existing Characterization documents need to be updated, new ones created, or for nothing to happen at all.

Continuous Characterization
Continuous Characterization keeps your Characterization up to date automatically, without any effort on your part.

Continuous Characterization keeps your Characterization up to date automatically, without any effort on your part. It's critical for making sure that your Task Demon Characterization stays in sync with your actual code, and it's shipping today for all users.

More coming soon

These two new features bring significant new versatility to the Task Demon way of building software, and also aid us in shipping more features faster. We'll be announcing a bunch of stuff over the coming weeks, so keep your eyes peeled, sign up for our newsletter if you haven't already, and head over to our Discord channel to hang out and ask us anything.

In the meantime, we have free trials available and the individual plan is only $15/month, so what are you waiting for?

Share Post:

Ready to transform your development workflow?

Join thousands of developers using Task Demon to build better software, faster. Start your free trial today.

Start Free TrialSchedule a Demo

How Task Demon transforms your workflow

Discover the key features that help developers build better software, faster.

AI-Powered Coding

Let intelligent agents help you implement features, fix bugs, and write tests.

Learn more

Task Triage & Planning

Automatically analyze and prioritize tasks with AI that understands your codebase.

Learn more

Privacy & Security

Run AI agents locally on your machine, keeping your code private and secure.

Learn more

GitHub Integration

Seamlessly connect with your repositories and sync issues automatically.

Learn more

Explore Task Demon

Discover more resources to help you get the most out of Task Demon.

Implementation Examples

See how Task Demon helps implement real-world features and fix bugs.

Learn more →

Video Demonstrations

Watch Task Demon in action with step-by-step demonstrations.

Learn more →

Feature Deep Dives

Explore detailed explanations of Task Demon's key features.

Learn more →

Why Task Demon?

Learn why teams choose Task Demon for AI-powered development.

Learn more →

Trusted by development teams

Teams using Task Demon are shipping faster, with higher quality code and less developer burnout.

    • Task Demon has revolutionized our development process. The AI triage system saves us hours of manual work categorizing and prioritizing issues.

      Alex Rivera
      CTO, TechFlow

    • The GitHub integration is seamless. It keeps everything in sync and the AI agents actually understand our codebase context when planning implementations.

      Jamie Chen
      Lead Developer

    • As an engineering manager, Task Demon gives me unprecedented visibility into our development process. The AI planning helps my team work more efficiently.

      Taylor Rodriguez
      Engineering Manager

    • The local agent using Claude Code is a game-changer. We get all the benefits of AI coding assistance while keeping our code private and secure.

      Sam Patel
      Security Engineer

    • The automated documentation feature alone is worth it. Task Demon generates comprehensive docs that actually stay up-to-date with our codebase.

      Jordan Kumar
      Senior Developer