Long prompts mix intent and configuration in one brittle text blob.
← Back to patterns
Inline & Contextual
Research PatternInline Prompt Controls
Parse the Prompt, Surface the Controls
What if your prompt parsed itself — every variable detected and turned into an inline control you click to change, right where it sits in the prose? Unlike a template you fill from a blank form, the prose stays the source of truth and the controls are extracted from it. No product ships this exact move yet (reliable parsing is the hard part); closest neighbours parse a fixed grammar out of plain text (Todoist Quick Add) or slot-fill references beside the prompt (Runway).
Metaphor
Rich Text Editor
Interaction
Inline, Contextual
Use case
Video Generation, Complex Prompt Authoring, Multi-Parameter Workflows
Failure mode
Config-in-Prose
Control mechanisms
Clickable Parameters · Popovers · Inline Editing
Reference examples
Todoist Quick Add · Runway Gen-4 Prompt + References
Interactive Example
The live side-by-side is the comparison.
This prototype already contains the contrast between chat and the pattern response. It stands as the primary comparison surface for this page.
The prompt is prose. It's also a control surface.
A long prompt fuses intent and configuration into one brittle text blob. Change the resolution or swap a single value, and you reread the whole thing, find the right words, and rewrite them in place without breaking the rest. Configuration buried in prose has no edit affordance of its own.
Inline Prompt Controls parse the variables out of the prose and make each detected value clickable where it sits — type and valid range on hover, structured options in place. The text stays the source of truth, but every parameter becomes directly editable instead of something you renegotiate with the whole sentence.
When editing one setting means rewriting the entire prompt, the prompt isn't an interface — it's a hostage situation.
DOPE Lens
How this pattern improves AI interaction quality.
Observability
Every parameter is visible inline, never hidden in syntax
Directability
Click any value to change it — the prompt is the control surface
Predictability
Structured options mean consistent, repeatable outputs
Explainability
Each detected parameter shows its type and valid range when hovered
Fit
Where it earns its place.
Use it
when
when
Prompts are long enough that editing one variable means rewriting the whole text.
Avoid
when
when
The prompt is intentionally freeform and the parser cannot detect stable fields.
Closest neighbours
No product ships this exactly — these come closest.
This is a research pattern — we describe a UX move that no shipping product implements cleanly today. The entries below are adjacent: they share the spirit of the pattern even if they take a different shape.
Todoist Quick Add
Doist
Adjacent — not generative AI, but the closest shipping parse. You type one line ("Meet with Ada tomorrow at 14 !30min p1"); Quick Add highlights the date, reminder and priority in place and mirrors each as a chip below, and the line stays the source of truth. Click a highlighted word to un-parse it. The parse covers a fixed grammar (dates, %labels, p1–p3, !reminders, +assignees), not free prose.
Visit Doist ↗
Runway Gen-4 Prompt + References
Runway
Adjacent — not exact. Runway pairs a free-text prompt with structured reference images and parameter controls rather than parsing typed syntax back into surfaced fields. Closer to slot-filling than to true parse-and-surface.
Visit Runway ↗
Research Frame
Why this pattern exists, and where it breaks.
Parse variables out of prose and surface them as inline controls.
Editing one setting means rereading and rewriting the entire prompt.
Bad parsing can invent structure and make the system feel unreliable.
Common questions
What people ask about this pattern.
When should I use the Inline Prompt Controls pattern?
+
Prompts are long enough that editing one variable means rewriting the whole text.
When should I avoid the Inline Prompt Controls pattern?
+
The prompt is intentionally freeform and the parser cannot detect stable fields.
What problem does Inline Prompt Controls solve?
+
Long prompts mix intent and configuration in one brittle text blob.
Why is chat the wrong fit for this?
+
Editing one setting means rereading and rewriting the entire prompt.
Related Patterns
Curated connections, not just neighbors.
Hand-picked relationships — what this pattern pairs with in practice, what it replaces, what it builds on, and what builds on it.
Often paired with
Property Panel
Inline Prompt Controls change values inside this prompt; the Property Panel holds stable settings across prompts.
Often paired with
Select & Transform
Inline Prompt Controls cover the variables the parser found; Select & Transform rewrites any other span you select as a diff.
Often paired with
Ghost Text Editor
Ghost Text Editor speeds up writing the prompt; Inline Prompt Controls turn each variable in it into a control you can click.