ForkFree

Prompts / Images and video

Image Brief With Exact Text

It gives you an image brief where quoted text is fixed and linked details such as light, shadow and reflections agree with each other. The closing checklist lets you spot spelling slips and inconsistencies quickly.

Worked in our testIt followed the structure and format: one paragraph in the right order, exact quoted text with surfaces, the 3:2 / 1K / photographic format, the heading, and five checks, with no extra commentary. It also linked light, shadows and reflection well. Two small flaws: it called the steam "backlit" although the window is at the side, and it invented cream colour for the italic and side text.

The prompt

Act as an art director who writes briefs for AI image generators.

Task: Turn my rough idea into one image brief that keeps all on-image text exact and all linked details consistent.

Steps:
1. Take the wording in {exact_text} and copy each item character for character inside quote marks, stating where it sits in the frame and on what surface.
2. Read {scene_details} and spot the details that depend on each other, such as weather, lighting, reflections and object positions. State how each one affects the others.
3. Write the brief as one clear paragraph in this order: subject, setting, text placement, lighting, camera. Use the format given in {image_format}.
4. Add a checklist of 5 things to verify in the finished image, covering spelling, placement and consistency.

Output: The brief paragraph first, then a heading "Check the image" with the 5 checklist items. No other commentary.

Change the parts in curly braces. Use it for packaging, signage, posters or infographics, where wrong spelling or mismatched details would ruin the image.

What happened when we ran it

We sent

Act as an art director who writes briefs for AI image generators.

Task: Turn my rough idea into one image brief that keeps all on-image text exact and all linked details consistent.

Steps:
1. Take the wording in Front of a tea tin: "MORNING SMOKE" in large cream capitals, and below it "Lapsang Souchong, 80g" in small italics. Side of the tin: "Blended in Bristol". and copy each item character for character inside quote marks, stating where it sits in the frame and on what surface.
2. Read A matte dark green tin sits on a wooden kitchen counter beside a white cup of steaming tea. Early morning light comes from a window on the left, so the shadows fall to the right. The steam should catch the light, and the cup's reflection should show on the glossy tin lid. and spot the details that depend on each other, such as weather, lighting, reflections and object positions. State how each one affects the others.
3. Write the brief as one clear paragraph in this order: subject, setting, text placement, lighting, camera. Use the format given in Landscape 3:2, 1K resolution, photographic style.
4. Add a checklist of 5 things to verify in the finished image, covering spelling, placement and consistency.

Output: The brief paragraph first, then a heading "Check the image" with the 5 checklist items. No other commentary.

The model replied

A matte dark green rectangular tea tin with a glossy lid stands slightly right of center, turned about 30 degrees so its front face and its left side face are both visible. To its right, close enough that its reflection reaches the lid, sits a plain white ceramic cup of freshly poured tea with a thin plume of steam rising from it. The scene is a warm wooden kitchen counter in early morning, with a softly blurred window on the left edge of the frame and an uncluttered kitchen background. Text appears only in three places. On the front face of the tin, in the upper half, is "MORNING SMOKE" in large cream capital letters, printed flat on the matte green surface. Directly below it on the same face is "Lapsang Souchong, 80g" in small cream italics. On the left side face of the tin, running horizontally in small cream type on the matte surface, is "Blended in Bristol". No other lettering, logos or labels appear anywhere. Soft, low, golden-white light enters from the window on the left, so every shadow falls to the right. The tin's shadow stretches across the counter toward the cup, and the cup's shadow falls to its own right. The left side face of the tin is the brightest vertical surface, and the front face is slightly softer. The steam is backlit and rim-lit by the window, glowing against the darker background on the right. The glossy lid shows a faint, distorted reflection of the white cup on its right half, nearest the cup, with a soft highlight from the window on its left edge. The matte body shows no reflections. The camera is at slightly raised eye level, about 20 degrees above the counter, with a 50mm lens at f/4. It is focused sharply on the tin's front text, with a gentle falloff behind the cup and window. The image is Landscape 3:2, 1K resolution, photographic style with natural color and fine surface texture in the wood grain and the tin's matte finish. ## Check the image 1. **Spelling:** The text reads exactly "MORNING SMOKE", "Lapsang Souchong, 80g" and "Blended in Bristol", with matching capitalization, the comma and no extra or garbled characters. 2. **Placement:** "MORNING SMOKE" is large and in capitals on the upper front face, "Lapsang Souchong, 80g" is small and italic directly beneath it, and "Blended in Bristol" is on the left side face, not the front or lid. 3. **Shadow direction:** All shadows, including the tin's, the cup's and the steam's faint cast, fall to the right, away from the window on the left. 4. **Reflection and finish:** The glossy lid shows a believable reflection of the white cup on the side nearest the cup, while the tin's body stays matte with no mirror-like reflections. 5. **Steam and light consistency:** The steam is visibly lit from the left and is clearly readable against the background. The left side face of the tin is the brighter of the two visible faces, and the text on it stays legible.

Run on Claude Sonnet 5.5, 08 Nov 2026. Automatic check: 4 out of 5.

Mentioned for

Nano Banana 2.1 Advanced, Nano Banana 2.1

Where it came from

Found on Web, shared by Wiro ML Team: Nano Banana 2.1 Advanced vs 2.1: 3 Sharp Tests.

This is our own version of the idea, written and tested by us. The original may differ.

More images and video prompts

Images and video Partly worked
Style Shortlist and Shot Plan for Code Animations

You get three contrasting style directions, a reasoned pick and a timed shot plan, all before spending time on code. Making the AI compare options first and tie the choice to the audience stops it defaulting to a generic look.

Found on GitHub, shared by XiaoyeCodes
Images and video Worked in our test
Motion-Capture-Friendly Video Prompt Writer

You get a video prompt that is written to suit motion tracking, plus a list of the risky moves it changed. It works because it puts the tracker's limits (clear limbs, one person, fixed camera) into the prompt before the video is made.

Found on GitHub, shared by blendi-remade
Images and video Worked in our test
Reusable Image Style Template Builder

You get a reusable style template, with the look locked in and the scene left as variables, plus a finished image prompt. Fixing the style traits separately from the content keeps results consistent across many images.

Found on GitHub, shared by VigoZhao