# AI use cases for XXG Portrait Rebuild Light (moskoo/xxg-portrait-rebuild-light) Repo: [moskoo/xxg-portrait-rebuild-light](https://github.com/moskoo/xxg-portrait-rebuild-light) | Use cases: 4 | Page: https://dedupedai.com/use-cases/moskoo-xxg-portrait-rebuild-light/ xxg-portrait-rebuild-light is a Python image-edit skill that relights existing portrait photos and repairs skin texture while keeping the person's identity, pose, and framing. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Relight a flat or weak backlit portrait: https://dedupedai.com/use-cases/moskoo-xxg-portrait-rebuild-light/relight-flat-backlit-portrait/ - Restore natural skin texture in a smoothed portrait: https://dedupedai.com/use-cases/moskoo-xxg-portrait-rebuild-light/repair-plastic-skin-texture/ - Give a portrait a film or camera-device look: https://dedupedai.com/use-cases/moskoo-xxg-portrait-rebuild-light/apply-film-or-device-look-to-portrait/ - Pull a multi-round AI portrait edit back to its root: https://dedupedai.com/use-cases/moskoo-xxg-portrait-rebuild-light/fix-drift-after-many-ai-edit-rounds/ ## How to relight a flat or weak backlit portrait with XXG Portrait Rebuild Light Repo: [moskoo/xxg-portrait-rebuild-light](https://github.com/moskoo/xxg-portrait-rebuild-light) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Rebuild key light, shadows, and highlights in an existing portrait without redrawing the person. **Who it is for:** A photographer or retoucher whose portrait was shot with weak backlight or flat indoor window light and needs the light to read as physically plausible. **The problem:** The source has flat lighting, crushed shadows, or highlights with no visible light source, which makes the portrait look unfinished. **The outcome:** An edited portrait with a director-led lighting change, with identity, pose, expression, and composition kept as in the source. ### Steps 1. Open https://github.com/moskoo/xxg-portrait-rebuild-light and read the skill instructions and the L and E recipe tables. 2. Give the agent the source portrait and state the light you want, for example soft natural window light or natural backlight. 3. Let the agent choose one L light code and one E exposure code, and check that the prompt uses the four core lines EDIT, LIGHT, SKIN, and AVOID. 4. Review the returned image for changed facial structure, pose, or framing. 5. If the edit misses the target, ask the agent for the single complete compact image-edit prompt the skill returns. ### Prompt to paste into your agent ```text Using the xxg-portrait-rebuild-light skill from moskoo/xxg-portrait-rebuild-light, relight the portrait at with soft natural window light, choosing one L light code and one E exposure code. Return the edited image, and verify that the face size, position, pose, and camera view match the source. ``` ### Caveat The README excerpt does not list install steps, so setup depends on how your agent loads the skill from the repo. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/moskoo-xxg-portrait-rebuild-light/relight-flat-backlit-portrait/ ## How to restore natural skin texture in a smoothed portrait with XXG Portrait Rebuild Light Repo: [moskoo/xxg-portrait-rebuild-light](https://github.com/moskoo/xxg-portrait-rebuild-light) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Replace plastic-looking skin with clean, low-contrast photographic microtexture while leaving lighting alone. **Who it is for:** A retoucher or creator whose portrait looks over-smoothed or waxy and needs skin fixed without changing the light. **The problem:** Heavy smoothing or AI generation has left plastic skin, and a full relight would change more than the skin. **The outcome:** A portrait with healthy skin microtexture and the source lighting, color, focus, and depth of field kept as they were. ### Steps 1. Open the repo at https://github.com/moskoo/xxg-portrait-rebuild-light and read the texture-only fidelity and S and P recipe sections. 2. Give the agent the portrait and say the request is texture-only, so the agent sets lighting, exposure, color, and atmosphere to their source defaults. 3. Ask the agent to choose one S skin scale code that fits the shot size, such as S1 for a medium shot or S2 for a high-resolution close-up. 4. Ask for P0 to keep the source diffuse and specular response, then check that pores, hair, and fabric detail are preserved and not exaggerated. ### Prompt to paste into your agent ```text Use the xxg-portrait-rebuild-light skill from moskoo/xxg-portrait-rebuild-light to run a texture-only skin repair on , with S1 scale and P0 finish. Return the edited image, and verify that the lighting, color, and focus match the source while the skin looks clean and not plastic. ``` ### Caveat The README says the skill keeps the source look for skin-only requests, but you still need to check the output yourself for over-sharpening or grain. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/moskoo-xxg-portrait-rebuild-light/repair-plastic-skin-texture/ ## How to give a portrait a film or camera-device look with XXG Portrait Rebuild Light Repo: [moskoo/xxg-portrait-rebuild-light](https://github.com/moskoo/xxg-portrait-rebuild-light) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Apply a color grade and capture response such as 35mm negative or instant film while keeping the view and focus. **Who it is for:** A content creator or stylist who wants a portrait to match a specific film stock or camera style for a series or campaign. **The problem:** A portrait needs a consistent retro or device-specific look, but the edit must not move the camera or change the focus. **The outcome:** A portrait with one color grade and one capture response applied, with viewpoint, perspective, crop, and depth of field left as in the source. ### Steps 1. Read the G color grade and D capture profile tables in the repo at https://github.com/moskoo/xxg-portrait-rebuild-light. 2. Choose the look you want, such as G7 for a restrained 1970s print response or D3 for 35mm color negative. 3. Tell the agent to keep viewpoint, perspective, crop, focal plane, and depth of field unless you ask for optical-restyle. 4. Compare the result with the source for skin tone and background color, and check that the grain is absent unless you asked for it. ### Prompt to paste into your agent ```text Using the xxg-portrait-rebuild-light skill from moskoo/xxg-portrait-rebuild-light, apply G7 color grade and D3 capture response to the portrait at , keeping viewpoint and depth of field. Return the edited image, and verify that the crop, focus, and skin tone match the source apart from the requested grade. ``` ### Caveat Grain, borders, and disposable-camera defects are opt-in in the README, so you must ask for them explicitly if you want them. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/moskoo-xxg-portrait-rebuild-light/apply-film-or-device-look-to-portrait/ ## How to pull a multi-round AI portrait edit back to its root with XXG Portrait Rebuild Light Repo: [moskoo/xxg-portrait-rebuild-light](https://github.com/moskoo/xxg-portrait-rebuild-light) | Difficulty: Advanced | Time: 45 minutes | Basis: From the README Remove accumulated color drift, noise, and seams after repeated AI edits by comparing against the first source. **Who it is for:** A creator or agent operator who has run several AI edit rounds on one portrait and now sees color shifts, seams, or banding. **The problem:** Each round of editing adds drift, sharpening, compression, and seams, so the image no longer matches the original subject. **The outcome:** A consolidated portrait that uses the first root base as the color and identity reference while keeping intended new changes. ### Steps 1. Keep the first source image as the root base and make it available to the agent alongside the latest edit. 2. Read the Q4 and Q5 quality repair entries in the repo at https://github.com/moskoo/xxg-portrait-rebuild-light. 3. Ask the agent to run Q5 to consolidate the edit against the root base, or Q2 and Q3 if you only see banding or seams. 4. Check that intended changes from earlier rounds remain and that the identity and colors match the root base. ### Prompt to paste into your agent ```text Use the xxg-portrait-rebuild-light skill from moskoo/xxg-portrait-rebuild-light to run a Q5 consolidation on , using as the root base. Return one consolidated image, and verify that identity and unchanged colors match the root while the banding and seams are gone. ``` ### Caveat The skill repairs only artifacts you can observe, so you still need to inspect each round to name the defects before asking for a Q code. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/moskoo-xxg-portrait-rebuild-light/fix-drift-after-many-ai-edit-rounds/