# Hodios paste pack: Image generation
Everything in Image generation from Hodios, the open prompt library by Hermes IDE: 7 entries, catalog 2026.1003.0.
Every entry is dedicated to the public domain under CC0 1.0. Copy, change and share them freely, no attribution needed.
Browse and search the library at https://hermes-ide.com/prompts
## How to use
Find an entry below and copy the text inside its block into ChatGPT, claude.ai or any chat. Replace each [PLACEHOLDER] with your own material. Personas, rules and styles work best as custom instructions or project instructions.
## Contents
- Image generation
- [Build a style kit for a series of generated images](#build-image-style-guide) (prompt)
- [Create a storyboard with image prompts](#create-storyboard) (prompt)
- [Write a video-generation prompt](#write-video-generation-prompt) (prompt)
- [Write an image-edit prompt](#write-image-edit-prompt) (prompt)
- [Write an image-generation prompt](#write-image-prompt) (prompt)
- [Write character consistency prompts](#write-character-consistency-prompts) (prompt)
- [Write product photo prompts](#write-product-photo-prompt) (prompt)
---
## Build a style kit for a series of generated images
`build-image-style-guide` · prompt · Image generation · https://hermes-ide.com/prompts/build-image-style-guide
Builds a reusable style kit (style statement, style tokens, prompt template, negative prompts, character sheets, consistency techniques, QA checklist) to keep generated image series consistent.
````markdown
A series of generated images drifts when each prompt is written from scratch: the palette shifts, the line weight changes, the main character gains a new face every time. Consistency comes from treating the look like a design system: a fixed vocabulary of style tokens reused word for word, a template where only the subject slot changes, references and seeds used deliberately, and a checklist for rejecting off-style results.
Build a style kit for this series:
[SERIES]
1. **Style statement:** 2 to 3 sentences describing the look in plain words, and 3 "not" statements that rule out the nearest wrong looks.
2. **Style tokens:** a fixed set of short phrases to reuse word for word in every prompt, grouped as medium and technique, line and texture, palette (named colours with hex values for reference), lighting, camera or viewpoint, composition rules, and mood. Mark which tokens are mandatory and which are optional.
3. **Prompt template:** a fill-in template with one slot for the subject and action and one for the setting, followed by the fixed style tokens in a fixed order. Keep the variable part at the start so the subject is not drowned out.
4. **Negative prompt or exclusions:** a reusable list for tools that support it, and, for tools that do not, the equivalent positive phrasing to put in the prompt.
5. **Recurring subjects:** for each character, mascot or object that appears in several images, a short sheet: a fixed description phrase (age, build, hair, clothing, colours, distinguishing features), what must never change, and the reference image to make first.
6. **Consistency techniques** for the tool: reference images and style references (for example Midjourney's style and character or object reference parameters, whose names depend on the version), fixed seeds for close variations, image-to-image or IP-Adapter-style references and LoRAs for Stable Diffusion, and, for chat-based image tools, keeping one conversation, re-attaching the reference and restating the full style block each time. Note which techniques are version-dependent and should be checked.
7. **Sample prompts:** 3 complete prompts from the template for different images in the series.
8. **QA checklist:** 6 to 10 yes-or-no checks to accept or reject an image (palette in range, line weight, character features, no stray text or watermarks, anatomy and hands, composition rule).
9. If the series description lacks the look or the use (no style hints, no idea where images go), ask up to three questions and stop.
- Describe styles by their qualities, not by naming living artists.
- Keep the style token block short enough that the subject still leads; about 25 to 40 words is a good target.
- Do not claim a technique guarantees identical results. Generative tools vary between runs and versions.
Markdown with the contract's sections as `##` headings, in order. Put the template, negative prompt and sample prompts in code blocks so they can be copied.
````
---
## Create a storyboard with image prompts
`create-storyboard` · prompt · Image generation · https://hermes-ide.com/prompts/create-storyboard
Creates a shot-by-shot storyboard with shot size, angle, movement, action, sound and timing, plus a consistent image prompt for every frame. Use when planning a video, animation or comic.
````markdown
A storyboard is where a story becomes pictures: which moments get a frame, how close the camera is, where the viewer's eye goes, and how one shot cuts to the next. Generated storyboards often fail at the basics: every frame is the same medium shot, the story beat in a frame is unclear, screen direction flips between shots, and the main character looks different in every image prompt. This prompt plans the shots like a director and writes the image prompts like a continuity supervisor.
Create a 8-frame storyboard.
[STORY]
1. **Concept:** the format, length, audience, and the one feeling or message the piece must leave, in 2 to 3 lines. If the story has no clear beginning, turn and end, propose one and label it.
2. **Beats:** choose the 8 moments that tell the story. Spend frames on turning points and emotion, not on transitions the viewer can infer. Use an establishing shot early unless the format calls for a cold open.
3. **Visual bible:** fixed descriptions reused word for word in every prompt: each character (age, build, face, hair, clothing, colours), key locations, the style tokens (medium, palette, lighting, aspect ratio). Characters are described by appearance, never by a real person's name.
4. **Storyboard:** for each frame give:
- number and beat;
- shot size (extreme wide, wide, medium, close-up, extreme close-up) and angle (eye level, low, high, overhead, over-the-shoulder);
- camera movement for video (static, pan, tilt, dolly, handheld) or panel size and placement for comics;
- the action, and where the subject sits in the frame;
- dialogue, caption, sound or music cue;
- duration in seconds for video (totals must match the target length), or page and panel for comics;
- the transition to the next frame (cut, match cut, dissolve, page turn).
Vary shot sizes to control rhythm, keep screen direction consistent (the 180-degree rule) unless a cross is intended, and use close-ups for the emotional peaks.
5. **Frame prompts:** one image prompt per frame, built from the visual bible plus that frame's shot size, angle, action and setting, formatted for the tool if one was given (for example Midjourney parameters at the end, or plain sentences for chat-based tools). Keep the character descriptions identical across frames.
6. **Continuity notes:** props, costumes, lighting and time of day that must carry over, and frames likely to need reference images to stay consistent.
7. If the story is too thin to choose beats (a single sentence with no event), ask up to three questions and stop. If 8 is outside 4 to 24, use the nearest bound and say so.
- Every frame must move the story forward or deliver an emotional beat; merge or drop frames that only repeat.
- No real, identifiable people, and no copyrighted characters in the prompts; describe original characters.
- Do not claim the image tool will keep characters identical. Recommend reference images or the tool's character-consistency features where they exist.
## Concept
## Visual bible
Characters, locations and style tokens, in a code block for copying.
## Storyboard
| # | Beat | Shot and angle | Movement or panel | Action and framing | Dialogue / sound | Duration or page | Transition |
## Frame prompts
Numbered code blocks, one per frame.
## Continuity notes
````
---
## Write a video-generation prompt
`write-video-generation-prompt` · prompt · Image generation · https://hermes-ide.com/prompts/write-video-generation-prompt
Writes prompts for AI video generators covering subject, action, camera movement, lighting, style, audio and consistency, with a shot-by-shot version for longer clips. Use with text-to-video tools.
````markdown
Video models read a prompt as a description of one continuous shot. They do best with one clear subject, one main action described with a precise verb, a specified camera behaviour, and a defined look. They struggle with several simultaneous complex actions, crowds interacting, fast hand movements, legible text, cause-and-effect physics and keeping a character identical across separate generations. Clips are short (often 5 to 10 seconds per generation, depending on the tool), so longer pieces are built from shots, and consistency comes from repeating the exact same descriptions of characters, wardrobe, setting and style in every shot and, where the tool supports it, from reference images or image-to-video starting frames. Some current tools also generate synchronised audio (dialogue, effects, ambience) from the prompt.
Write a video-generation prompt for this idea.
[IDEA]
Target tool: generic
Total duration: 8 seconds
1. **Interpretation:** two or three lines on the video you are aiming for, the choices you made where the idea was open, and whether it fits in one shot. If the idea is too open to film (for example "something epic for my brand"), ask up to three questions (subject, setting, purpose or mood) and stop.
2. **Prompt** (one shot): write a single paragraph in this order, with concrete visual words:
- **Shot and camera:** shot size, angle, lens feel, and one camera movement (static, slow push in, pull out, pan, tilt, tracking alongside, orbit, crane up, handheld), with its speed;
- **Subject:** who or what, with fixed visual identifiers (age range, build, hair, clothing, colours);
- **Action:** one main action with a precise verb and its pace, beginning to end within the shot;
- **Setting:** place, time of day, weather, background activity kept simple;
- **Light:** source, direction, quality and colour;
- **Style:** live-action cinematic, documentary, animation style described by technique, film stock or grade, frame rate feel (slow motion, real time);
- **Audio** (only for tools that generate sound): ambience, effects, and any short line of dialogue in quotes with who says it.
3. **Shot list:** if 8 seconds is longer than one generation in generic (assume about 5 to 10 seconds per shot if unsure, and say so), split it into shots in a table: shot number, duration, shot size and camera, action, transition (cut, match cut, continuous), and a full standalone prompt for each shot that repeats the consistency block. If one shot suffices, write "Single shot".
4. **Consistency block:** a reusable paragraph describing each recurring character, outfit, setting and visual style in fixed wording to paste into every shot; recommend generating a reference image first and using image-to-video or the tool's reference or character feature if it has one.
5. **Settings:** aspect ratio for the use (16:9 for widescreen, 9:16 for vertical social, 1:1 or 4:5 for feeds), duration per clip, and any settings the tool commonly offers (motion strength, seed reuse for consistency, resolution). Note that settings and limits change between versions and tell the user to check their tool's documentation.
6. **Variations:** two alternative prompts that change one decision each (camera, light or style) and what each changes.
7. **Troubleshooting:** three fixes specific to this video for common failures (subject morphing, too much or too little motion, the camera ignoring instructions, warped hands or faces, unwanted text).
- Keep each shot prompt focused: one subject focus, one main action, one camera movement. Move extra actions into separate shots.
- Do not ask the model to render readable text in the frame; add text in editing instead and say so.
- Do not write prompts that depict real, identifiable people (including public figures) doing or saying things they did not do, sexualised content of real people or any minors, or footage designed to pass as real news, evidence or a real brand's advertising.
- Do not imitate copyrighted characters or a living director's signature style by name; describe the visual qualities instead.
- Phrase prompts positively ("an empty street") rather than relying on negatives, unless the tool has a separate negative prompt field.
## Interpretation
## Prompt
In a code block, ready to paste.
## Shot list
Table, then one code block per shot; or "Single shot".
## Consistency block
In a code block.
## Settings
## Variations
## Troubleshooting
````
---
## Write an image-edit prompt
`write-image-edit-prompt` · prompt · Image generation · https://hermes-ide.com/prompts/write-image-edit-prompt
Writes precise AI image-edit instructions for inpainting, background changes, style transfer or object removal that state what must not change, with mask guidance and checks. Use before an edit.
````markdown
Image edits go wrong in predictable ways: the model "improves" things it was not asked to touch (faces change, text garbles, the framing shifts), the inserted element does not match the light, perspective or grain of the original, removed objects leave smears or ghost shadows, and a single prompt that asks for several changes does none of them well. Good edit instructions name the one change, describe exactly what must stay the same, specify how the new content should match the existing image (light direction, colour temperature, shadows, perspective, texture), and break complex edits into passes. Mask-based tools add another rule: the prompt describes what should fill the masked area, not the whole picture.
Write generic edit instructions.
[IMAGE_DESCRIPTION]
[DESIRED_CHANGE]
1. **Edit plan:** restate the change in one line and split it into passes if it involves more than one change (for example remove the bin first, then change the sky). If the description of the image is too thin to know what must be preserved or how the light falls, ask up to three questions and stop.
2. **Preserve list:** everything that must not change: identity and facial features, pose, framing and crop, the product or logo, any text, the lighting direction and colour temperature, background elements, image style and grain. Be specific to this image.
3. **Prompts:** one per pass, formatted for generic:
- **instruction:** plain sentences: the change first ("Replace the overcast sky with a warm sunset sky with soft pink and orange clouds"), then a matching instruction ("Adjust the light on the building to warm, low sun from the left to match"), then "Keep everything else exactly the same:" followed by the preserve list.
- **inpainting:** a prompt describing only what fills the masked area, in the same style and light as the surrounding image, plus a negative prompt if the interface has one.
- **region-editor:** a short prompt for the selected region that describes the new content and how it matches its surroundings, with a note on which area to select.
- **generic:** a clear instruction paragraph, then "Preserve:" and the list.
For removals, describe what should be behind the removed object (continuing pavement, the rest of the wall pattern), not the object. For background changes, keep the subject's edges, add contact shadows and match light direction and colour temperature. For style transfer, keep composition and identity and state how strong the stylisation should be.
4. **Mask and settings:** for mask-based tools: what to mask (include shadows and reflections of removed objects; mask slightly beyond the edges for clean blending, but not into areas that must stay), and starting values for denoising or strength (about 0.3 to 0.5 for subtle changes, 0.6 to 0.8 to replace content, higher only to generate something new). For others: any useful settings, and to start from the original image each time rather than re-editing an already edited result when quality drifts. Tell the user settings vary by tool.
5. **Checks:** a checklist for reviewing the result: preserved items unchanged (compare side by side at 100 percent zoom), light and shadow direction consistent, edges clean, no repeated textures or artefacts, text and logos intact, and what to try if each check fails.
- One change per pass. Do not add improvements the user did not ask for.
- Do not write instructions to remove watermarks or credits from images the user does not own, to alter identity documents or evidence, or to place a real, identifiable person into a compromising, sexual or deceptive scene.
- If the edit would make a product photo misrepresent the real product (colour, size, features) for a listing, flag it.
## Edit plan
## Preserve list
## Prompts
One code block per pass, labelled Pass 1, Pass 2.
## Mask and settings
## Checks
A checklist.
````
---
## Write an image-generation prompt
`write-image-prompt` · prompt · Image generation · https://hermes-ide.com/prompts/write-image-prompt
Turns a rough idea into a detailed image-generation prompt (subject, composition, lighting, style, lens) in the chosen tool's syntax, with settings and variations. Use before generating an image.
````markdown
Weak image prompts are either too thin ("a cat in space") so the model fills every gap with its defaults, or stuffed with filler ("masterpiece, 8k, trending, ultra detailed") that current models mostly ignore. Strong prompts describe the picture a photographer or illustrator would plan: the subject and what it is doing, the setting, the framing and camera, the light, the medium and style, the palette and mood. They also respect how each tool reads text: some take parameters, some take a separate negative prompt, some read only plain sentences.
Write a generic prompt at aspect ratio 1:1 for this idea:
[IDEA]
1. **Interpretation:** in 2 to 3 lines, say what image you are aiming for and any choices you made where the idea was open (subject details, setting, style). If the idea is so open that the result could go in very different directions (for example "something cool for my brand"), ask up to three questions and stop.
2. Build the description in this order, using concrete visual words:
- **Subject:** who or what, appearance, pose or action, expression;
- **Setting:** place, time of day, weather, background elements;
- **Composition:** shot size (close-up, medium, wide), angle (eye level, low, overhead), subject placement, depth of field, negative space for text if the use needs it;
- **Light:** source, direction, quality and colour (soft window light from the left, golden hour backlight, hard noon sun);
- **Medium and style:** photograph, oil painting, flat vector, 3D render, ink, and so on, described by technique and era rather than by naming a living artist;
- **Camera or rendering details** where they help: lens focal length, film stock look, aperture for photographs; brush or line quality for illustration;
- **Palette and mood.**
3. Format it for generic:
- **midjourney:** one descriptive prompt in natural language, most important elements first, then parameters at the end: `--ar 1:1`, and where useful `--no` for unwanted elements, `--style raw` for a more literal photographic look, or `--stylize` to tune how much the model's own aesthetic applies. Parameter names and ranges change between versions; tell the user to check them for their version.
- **stable-diffusion:** a positive prompt and a separate negative prompt. Use concise comma-separated phrases; mention that attention weights like `(golden light:1.2)` work in common interfaces such as AUTOMATIC1111 and ComfyUI, and that newer models (SD3, FLUX) follow full sentences better and may ignore or not support negative prompts. Give suggested width and height in multiples of 64 that match the ratio near the model's native resolution, plus typical steps and guidance (CFG) values as starting points.
- **chat-based:** plain, complete sentences, as you would brief an illustrator. No parameter syntax, no weights. State the orientation and aspect ratio in words, and tell the user to pick the matching size setting if the tool has one. Phrase exclusions positively ("an empty beach") because there is no negative prompt field. Put any text that must appear in the image in quotes, exactly as it should be spelled, and keep it short.
- **generic:** a clear natural-language paragraph, then an "Avoid:" line, then the aspect ratio.
4. Give 2 variations that change one decision each (composition, light or style) and say what each changes.
5. Give 3 tuning tips specific to this image: what to change if the result is too busy, wrong in mood, or misses a detail.
- Do not add filler quality tags ("masterpiece", "8k", "best quality") unless the tool is stable-diffusion with a model known to respond to them, and say so if you do.
- Do not name living artists as a style to copy. Describe the stylistic qualities instead.
- Do not write prompts for realistic images of real, identifiable people in false or sexual situations, or for images that imitate a real organisation's branding to deceive.
- Keep the main prompt under about 75 words for midjourney and stable-diffusion; detail beyond that is often ignored.
## Interpretation
## Prompt
The prompt in a code block, ready to paste. For stable-diffusion, two code blocks labelled Positive and Negative.
## Settings
Aspect ratio and any tool settings, or "none needed".
## Variations
Two code blocks, each with a one-line note.
## Tuning tips
Idea: "cosy reading nook for a bookshop's Instagram", tool generic, 4:5.
Prompt: "A cosy reading nook in a small independent bookshop on a rainy afternoon. A deep green velvet armchair beside a tall wooden bookshelf, a knitted blanket and a steaming mug of tea on a side table. Soft warm lamplight from the left, rain on the window behind, shallow depth of field. Photograph, 35 mm lens, muted warm palette of green, amber and cream, calm and inviting. Empty space at the top for a caption." Avoid: people, visible logos, text. Aspect ratio 4:5.
````
---
## Write character consistency prompts
`write-character-consistency-prompts` · prompt · Image generation · https://hermes-ide.com/prompts/write-character-consistency-prompts
Builds a character sheet and prompt kit that keeps one character recognisable across image generations, with locked traits, outfits, poses, expressions, a reference strategy and a drift checklist.
````markdown
You are a character designer who works with image generators for picture books, comics, storyboards and brand mascots. Image models have no memory of a character between generations, so the face, proportions and outfit drift unless you constrain them. What works: describing the character in the same exact words every time (an identity block reused verbatim), choosing a few distinctive, drawable traits instead of many vague ones, generating a canonical reference (a turnaround or character sheet) first and then using the tool's image-reference or character-reference feature, keeping style words separate from identity words, and checking each output against a fixed list before accepting it.
[CHARACTER_DESCRIPTION]
1. If the description is too thin to draw a recognisable person (for example only "a knight" or "a cute girl"), ask up to three questions and stop. Otherwise fill single missing details (build, eye colour, height) with assumptions that fit, and list them. If no style is given, choose one that fits the stated use, write it as a separate style block, and name two alternatives; because identity and style are kept apart, the user can swap the style block without touching the rest of the kit.
2. Write the character sheet: name, age, height and build, face shape, skin tone, eyes, hair (colour, length, style), three to five signature traits that make the character recognisable at thumbnail size (a scar, a colour, an accessory, a silhouette), and the personality to convey through posture.
3. Write the identity block: one compact paragraph of 40 to 70 words, in the order models weight most (subject, face and hair, signature traits, build), to paste unchanged into every prompt. Keep style words out of it.
4. Define two to four outfits as named, reusable blocks with the colours fixed.
5. Write a pose and expression set: a turnaround sheet prompt (front, three-quarter, side, back on a plain background) to generate the canonical reference first, then six to eight scene prompts combining the identity block, one outfit, a pose, an expression, the setting and the style block.
6. Write the reference strategy: generate and pick the canonical image first; then use reference images, character or image-reference features, fixed seeds where supported, inpainting for fixes, and, for long projects, a fine-tuned model or adapter trained on approved images. Give general guidance that applies to any tool and short notes for the tool if one is named, telling the user to check current parameter names in its documentation.
7. Write a negative prompt (or "avoid" list for tools without one) targeting the drift this character is prone to.
8. Write a drift checklist to accept or reject each output.
- The identity block never changes between prompts. Variation comes only from the outfit, pose, expression, setting and style slots.
- Prefer a few distinctive, visual traits over long lists; models blur long descriptions.
- Do not invent tool parameters or version-specific flags. If you are unsure a feature exists in the named tool, say so.
- Do not base the character on a real, identifiable person's likeness without saying the user needs that person's consent, and do not design characters that copy a trademarked character.
- Keep the kit tool-neutral unless a tool is named.
## Character sheet
A table of attributes, then signature traits as bullets. Assumptions, including the chosen style if none was given.
## Identity block
A code block with the reusable paragraph.
## Outfits
Named code blocks, then the style block as its own code block.
## Pose and expression set
The turnaround prompt, then numbered scene prompts as code blocks.
## Reference strategy
Numbered steps, then tool notes.
## Negative prompt
A code block.
## Drift checklist
A checklist.
````
---
## Write product photo prompts
`write-product-photo-prompt` · prompt · Image generation · https://hermes-ide.com/prompts/write-product-photo-prompt
Writes product-photography image prompts covering set, lighting, angles, props and lifestyle scenes in one consistent brand look, with variants for listings and ads. Use for e-commerce imagery.
````markdown
Product imagery sells when it is accurate, consistent and shows the product in use. An image model will happily invent a product that looks better than the real one, with a different shape, a misspelled logo or the wrong colour, which leads to returns, bad reviews and listings removed for misrepresentation. The reliable approach is to use generation for sets, backgrounds, props and lifestyle scenes, and to keep the real product faithful by supplying a reference photo (image-to-image, reference or product-placement features) or compositing the real product into the generated scene. A shot set needs the standard e-commerce angles plus lifestyle and ad formats, all sharing one lighting recipe, palette and surface vocabulary so the store looks like one brand.
Write generic product-photo prompts for:
[PRODUCT]
Brand style: [BRAND_STYLE]
1. **Product accuracy:** list the attributes that must stay faithful in every image (shape and proportions, materials and finish, exact colours, logo and label placement, size relative to a hand or common object). If the product description lacks shape, colour or materials, ask up to three questions and stop. Recommend supplying a clean reference photo of the real product and using the tool's image reference or product-placement feature, or compositing the real product photo, for any image that shows the product itself.
2. **Look:** the lighting recipe (for example large soft key light from the upper left, white bounce fill, gentle shadow; or hard sunlight with crisp shadows), the palette, surfaces and backgrounds, prop vocabulary, camera and lens feel, and mood. If brand style is blank, propose one that suits the product and customer and say why.
3. **Shot set:** prompts for each shot below, in a table with use, aspect ratio and notes, followed by each prompt ready to paste:
- **Hero on white:** the product alone on a pure white seamless background, filling most of the frame, soft even light, no props, no text: the usual marketplace main image;
- **Angles:** front three-quarter, side or back, and top-down, on the brand background;
- **Detail:** a close-up of the material, texture or key feature;
- **Scale:** the product with a hand or a familiar object for size;
- **Lifestyle:** two scenes of the product in use by the target customer, in a setting that fits the brand;
- **Ad variants:** a square 1:1 and a vertical 9:16 composition with clear empty space for a headline, and a 4:5 feed image;
- **Seasonal or campaign** (one): the same look adapted to a season or launch.
4. **Style suffix:** a reusable block of lighting, palette, surface and camera wording to append to any future prompt for consistency.
5. **Accuracy and compliance:** what to check in every output against the real product (logo spelling, colour against the real item under daylight, proportions, number of parts), and a reminder that many marketplaces require main images to show the actual product accurately on a pure white background, often with rules about props, text and how much of the frame the product fills; tell the user to check their marketplace's current image rules.
- Do not let prompts change the product's design, add features or accessories that are not included, or show results the product cannot deliver; props must be clearly separate from what is sold, or noted as "not included".
- Keep prompts free of filler quality tags ("8k", "masterpiece") unless the tool is stable-diffusion and the user's model is known to respond to them.
- Do not render text, prices or claims in the image; add them in design software.
- Do not imitate another brand's trade dress, logos or recognisable campaign imagery, and do not use real people's likeness without permission.
## Product accuracy
## Look
## Shot set
Table: Shot | Use | Aspect ratio | Notes, then one code block per prompt.
## Style suffix
In a code block.
## Accuracy and compliance
A checklist.
````