Pexo

THE PEXO PROMPT LIBRARY / TEMPLATE

AI Prompt Enhancer Template: Improve a Rough Brief

Improve a rough visual brief by filling the missing information yourself. This page is an editing worksheet, not an automated prompt enhancer or a promise of better output. Fill the worksheet, copy the completed prompt into the relevant workflow, and review the actual result. The examples below are editorial drafts, not observed outputs.

A rough pencil creative sketch beside a clearer multi-frame shot plan and organized color references
Illustrative concept image made for this page · not a tested product output.

Prompt Improvement Worksheet

Replace each bracket with your own information. Keep unknown facts out of the brief.

What would show that a revised prompt is better?

Use the same authorized source and one stated pass condition for both versions. If the rough brief omits a logo-preservation rule, add only that rule, then compare the two actual outputs for logo fidelity and the requested main action. Do not infer success from a longer prompt or from the worksheet itself; keep both results if a test is run.

Prompt text
Rough brief: [paste your own text]
Audience and purpose: [who and why]
Approved source assets: [images, script, facts]
Main action or edit: [one observable change]
Style only where useful: [light, camera, sound]
Must preserve: [identity, logo, claims, exact wording]
Review criteria: [what would count as correct]

Which brief fits your source?

Brief 1: Rough-brief diagnostic worksheet. Brief 2: Approved-mug product-video revision.

Selection rule: Use the worksheet to locate missing source and review fields; use the mug example to add protected handle, color and logo details.

Source-to-instruction relationship: A longer brief is not automatically better; the revision should change one observable criterion while using the same source.

If the draft misses the brief

Visible signal: The revised text adds adjectives but still permits a fabricated handle or logo.

Next revision: Replace style filler with an explicit handle-and-logo preservation rule and compare both outputs only if a real test is run.

Keep the source and review together

Source to retain: both prompt versions and the same approved mug photos if a comparison is run. After an actual run: save the exact prompt version, resulting file and a review note stating whether the revised version improves handle and logo fidelity without relying on prompt length. Until that run exists, leave the result and review fields unfilled. The Gallery cases below are separate published projects and cannot complete this record.

Before and After Example

Compare the rough and revised prompts. Add only details that affect the viewer’s understanding or protect source accuracy. If a condition is unknown, ask for it rather than inventing it.

Example 1: Before: Make a cool video for my mug.

Prompt text
Use my approved ceramic-mug photos to plan a short product video for an online shopper. Open on the actual lid, show a hand lifting the mug by its real handle, and end on the full product profile. Keep the color and logo unchanged; leave space for approved copy.

Example 2: Before: Make this picture better.

Prompt text
Edit my authorized product image. Reduce background clutter and soften the light while keeping the package shape, label, camera angle and color unchanged. Show a preview so I can compare it with the original.

These are written example prompts; they are not screenshots or measured outputs.

Add Missing Constraints

Add only approved assets, exact facts and a visible review criterion. If a fact is unknown, ask for it rather than guessing.

Keep a one-change review log

Use four fields: original goal, revised clause, observed result and next change. For the mug brief, first add the approved mug photos and a lid-opening shot; after a real run, record whether the lid and logo match the source. Leave the result field blank until a result exists. If the lid fails, revise that clause before changing lighting or duration, so the next comparison has one clear cause.

Test One Revision

Add constraints that protect source accuracy: approved assets, facts to keep exact, edits that are allowed and a reviewable end frame. Test the revised brief once, compare the actual result with the original goal and change one unclear instruction. More words alone do not improve a prompt; no example here is a tested output.

Use the Brief in Pexo

After revising the brief, choose the matching live workflow: Pexo Image Generation for a still or Pexo AI Storyboard Generator for a video shot plan. Supply authorized assets, compare the first result with the constraints above and request one targeted change. This worksheet itself does not enhance prompts automatically.

Privacy and Usage Notes

Use only references you have permission to submit. Remove unnecessary personal or customer information. Before sending sensitive material to any service, review its current terms and privacy policy; this worksheet makes no claim about storage, deletion or account conditions.

Test one revision at a time

Use this brief when: Use the worksheet when a rough brief lacks audience, source assets or a reviewable result.

Acceptance check: Compare the original and revised text line by line: each added fact should come from an approved source and each new constraint should solve a named problem.

If it fails: If the output still fails, change one constraint and keep the previous version for comparison.

Compare one change against the same input

Save the rough and revised prompts, use the same source asset, model and settings, and change only one constraint at a time. Before running either version, state a pass condition: for the poster example, the approved headline must remain exact, the subject must remain visible and the requested size must be usable. Keep both outputs and mark the first failed condition rather than calling the longer prompt better by default.

OpenAI's model-optimization guide recommends representative test inputs and evaluation criteria when iterating prompts. That is a general testing method, not evidence that this worksheet or Pexo ran the example. This page offers a review template; without paired outputs it cannot claim a measured improvement.

AI Prompt Enhancer FAQ

Does this worksheet rewrite my prompt automatically?

No. It shows the fields to add and examples of a careful manual rewrite.

Which constraint should I add first?

Start with the approved source facts and the detail the output must not change; then add the visible action and review goal.