Write the work once.
Queue prompts after each other with the sequence marker. One task becomes a script instead of a babysitting session.
Research the topic ⏭︎ Draft the outline ⏭︎ Write the first pageWrite a book outline about {Subject}⏭︎Draft chapter prompts⏭︎Summarize the whole plan
prompt outline
Prolifica Automation Studio
One task becomes many prompts. A Chrome beta that runs in your browser: use the sequence marker, multiply variables with braces, import data, choose repeat rules, and let Magic Planner turn a goal into a runnable prompt sequence.
The product story
The human pain is simple: hours spent babysitting AI tools. Prolifica turns that patience tax into a visible automation grammar.
The promise is not vague productivity. It is controlled multiplication: make the sequence, define the subjects, preview the run, then let the browser do the repetitive sending.
Sequence ready
Sequence ready
Sequence ready
Queue prompts after each other with the sequence marker. One task becomes a script instead of a babysitting session.
Research the topic ⏭︎ Draft the outline ⏭︎ Write the first pageWrap a subject in braces and Prolifica expands it into generated prompts for each value.
Write a book about {Love, Apples, Cherries}Attach data to a variable or import a whole sequence from rows, then preview the generated prompts before running.
{Subject} = your imported subject listRepeat only the prompt that has variants, or repeat the whole sequence for every subject.
Prompt only: 10 prompts / Whole sequence: 18 promptsMagic Planner breaks a large goal into smaller subtasks, then imports the plan as a sequence.
Goal: write a book → research → outline → draft → pagesSyntax variations
The same engine has several shapes: hand-written sequences, inline variants, file-backed variables, and imported prompt rows.
Write me a book about love ⏭︎ Write me a book about apples ⏭︎ Write me a summary about bothWrite me a book about {Love, Apples, Cherries}Write me a book about {Subject}prompts.csvAutomation Engine
The engine is built from a few clear moves: split prompts into order, mark what changes, bring in data, choose the repeat rule, then run locally in the browser.
Use the sequence marker to split one instruction into ordered prompts that run one after another.
Use braces for subjects, audiences, tones, products, or any repeatable variable.
Connect imported values to variables, or turn imported rows into a sequence of prompts.
Auto mode waits for completion. Timer mode covers surfaces where completion cannot be read.
Planner creates subtasks and lets the user import the plan back into Prolifica as a sequence.
Prolifica runs locally in the browser and does not access the user’s prompt sequences or work.
Run variations
These are the concrete demo shapes: a 60-prompt outline run, five outlines generated from subject inputs, and a Planner goal that becomes a sequence.
A long outline run from the demo: prompt sequence, automatic sending, finished in about 25 minutes.
Subjects from data inputs: Productivity Hacks, AI Ethics, Future of Work, Creative Writing Tips, Business Innovation.
“Outline a series of blog posts on sustainable tech” becomes an editable plan, then a sequence.
Repeat and conversation rules
Big runs should explain themselves: how many prompts will run, why that number changed, and whether the whole sequence or only one variant prompt repeats.
Only the prompts with variants expand. The rest of the sequence runs once.
The full sequence repeats for each subject. This is the default behavior for variant runs.
Each subject can open in its own conversation so big runs do not collapse into one thread.
Run control
The run count, sending mode, and repeat behavior need to be legible before any prompt starts moving.
Best when completion can be read
Best when completion cannot be read
Magic Planner
Planner creates an action plan based on a goal and a number of steps. The important move: the plan can be imported as a Prolifica sequence and run inside the engine.
Outline a series of blog posts on sustainable tech.
Browser boundary
Prolifica plugs into the user’s existing AI services, stores prompt sequences locally on their computer, and does not access those prompts or work.
Chrome beta
Next in the workshop