Whiteboard Canvas
Give a team an infinite canvas for notes, shapes and arrows they can move around together.
What it adds
A pannable, zoomable canvas holding sticky notes, basic shapes, text and connectors, editable by several people at once.
What your agent is told to do
5
What your agent is told to do
5-
1
Ship a small, coherent object set first: sticky note, rectangle, ellipse, free text, and a connector between two objects. A general-purpose diagramming engine is not the minimum version and should be deferred.
-
2
Reuse the app's existing realtime and presence layer. If Presence Indicators already exists, show collaborators on the canvas through it rather than opening a second live channel.
-
3
Make every object move, resize and delete with both pointer and keyboard, and give undo the same coverage. If the app has Undo Actions, canvas edits go into that history rather than a private stack of their own.
-
4
Persist continuously as people work. Reuse the app's Autosave behaviour if it exists, and make the saved state authoritative so a reload restores the board exactly.
-
5
Do not attempt pixel-perfect freehand drawing, image import and a template gallery in the first pass. Name them as deferred so the board that ships is finished rather than three things that are each nearly working.
Edge cases it handles
8
Edge cases it handles
8- Two people dragging at once must both keep their work. Merge concurrent edits per object and per property, so one person's move does not overwrite another's resize or wipe a note they were typing into.
- Pan and zoom must stay smooth as the board grows into the thousands of objects. Render only what is in view, and decide early what happens past that ceiling rather than letting the board grind to a halt.
- Connectors are attached to objects, not to coordinates. Moving a shape drags its connectors with it, and deleting a shape must resolve the connector deliberately rather than leaving an arrow pointing at nothing.
- Mouse, touch and trackpad all mean different things by the same gesture. Pinch, two-finger scroll and drag must behave consistently, and a trackpad scroll must not zoom when the user meant to pan.
- An exported snapshot must match what is on screen, including fonts. If a font is unavailable at export time the text reflows and the export no longer resembles the board.
- A dropped connection must not lose work. Queue edits locally, reconcile on reconnect, and tell the user plainly that they are offline rather than silently discarding changes.
- Boards are user-generated content. Decide who can view, who can edit, and what a shared link exposes before shipping sharing at all.
- Objects need readable names or labels for anyone not using a pointer, and an accessible list view of the board's contents is a better fallback than an empty canvas.
Definition of done
9
Definition of done
9- Notes, shapes, text and connectors can be created, moved, resized and deleted by pointer and by keyboard.
- Simultaneous edits by several people merge without one person's change erasing another's.
- Connectors stay attached when their shapes move and resolve deliberately when a shape is deleted.
- Pan and zoom stay responsive at the documented object ceiling.
- Board state persists continuously and a reload restores it exactly.
- An exported snapshot matches the on-screen layout and fonts.
- Presence and collaboration run through the app's existing realtime layer.
- The feature matches the existing design system.
- No existing functionality is broken.
Related features
Survey Builder
Survey Builder
Build a multi-question survey, send it out, and read all the answers in one place.
What it does
An authoring tool for multi-question surveys, a respondent flow, and a results view with export.
How it works
- 1 Ship a small, honest set of question types first: single choice, multiple choice, short text, long text, and a rating scale. That covers most real surveys. Defer matrix grids, file uploads, and calculated fields rather than half-building them.
- 2 Render the respondent flow with the app's existing Multi-Step Form and Conditional Form Fields, and reuse Form Field Shell for the inputs, so a survey looks and behaves like every other form in the product.
- 3 Give surveys the app's existing Draft and Publish States. Editing a live survey while responses are arriving is the main way survey data gets corrupted, so publishing must be a deliberate act with a clear boundary.
Copy the prompt
No account needed
Add this feature to my app:
https://addthisfeature.com/x/survey-builder
Automation Builder
Automation Builder
Let customers wire a trigger, some conditions and an action so routine work runs itself.
What it does
A visual builder where a customer connects a trigger to conditions and actions, plus the runtime that executes it and a log of what happened.
How it works
- 1 Ship a fixed, curated set of triggers, conditions and actions drawn from what the app already does. A general-purpose scripting environment is not the minimum version, and offering one you cannot support is worse than offering ten steps that work.
- 2 Run automations through the app's existing background job system. If Background Job Dashboard or Scheduled Job Monitor exists, automation runs appear there rather than in a separate queue nobody watches.
- 3 Make the run log the centre of the feature, not an afterthought. A customer whose automation did nothing needs to see which run fired, what data entered each step, and where it stopped.
Copy the prompt
No account needed
Add this feature to my app:
https://addthisfeature.com/x/automation-builder
Quiz Builder
Quiz Builder
Build a scored quiz with branching questions and a result page that reflects the answers.
What it does
An authoring tool for multi-question quizzes with scoring rules and answer-based branching, plus the taking experience and a result page.
How it works
- 1 Reuse the app's existing form machinery for the taking experience. If Multi-Step Form and Conditional Form Fields exist, the quiz runner is those features driven by the quiz definition, not a second form engine.
- 2 Support a small set of question types that actually score: single choice, multiple choice, and short text with an exact or pattern match. Free-text grading and media questions are a separate scope and should be deferred.
- 3 Give the author a way to see the quiz as a taker would, including a specific answer path, before it is published. A branching quiz cannot be checked by reading the editor.
Copy the prompt
No account needed
Add this feature to my app:
https://addthisfeature.com/x/quiz-builder
How it works
-
1
Copy the link
Grab the Markdown instruction URL for this feature.
-
2
Give it to your AI
Paste it into Claude Code, Cursor, v0, Lovable — whatever you build with.
-
3
It inspects, then implements
Your agent reads your existing app first, then adds the feature to fit it.
Works with your stack
These instructions are written to adapt. They tell the agent to detect your framework, match your existing design system, and reuse what you already have — rather than assuming a particular stack.
Need it tighter than that? Customize the feature and tell it exactly what you're running.