image-reference-edit
/image-reference-edit
Use when the task is to edit an image while preserving the approved subject and unchanged regions.
- Author
- voyager67 ↗
- Repository
- Agentic-Skill-Garage ↗
- Licence
- Apache-2.0
- Last repository change
- 07.11.2026
Assessment and security
- Files in this package
- 1
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 07.11.2026, without references to retired models or APIs.
Apache-2.0, otwarta: można używać i zmieniać, z zachowaniem licencji i informacji o zmianach.
What it does and when to use it
Use when the task is to edit an image while preserving the approved subject and unchanged regions.
Original author description
Use when the task is to edit an image while preserving the approved subject and unchanged regions.
How it works
- Outcome
- Inputs and prerequisites
- Procedure
- Quality gates
- Deliverable
- Boundaries
SKILL.md preview
--- name: image-reference-edit description: Use when the task is to edit an image while preserving the approved subject and unchanged regions. license: Apache-2.0 --- # Image Reference Edit ## Outcome Edit an image while preserving the approved subject and unchanged regions. The result must be traceable to supplied inputs and pass the checks below. ## Inputs and prerequisites - A visual brief, target dimensions, output format, reference roles, approved identity constraints, and a review bar. - ComfyUI or another explicitly configured generation/editing tool, plus Pillow, GIMP, Krita, or ImageMagick when installed. - Confirm the target application's version and the actual capabilities available in this environment. ## Procedure 1. Separate subject, composition, camera, lighting, palette, text, and negative constraints in the brief. 2. Choose a tool that supports the required control; prefer deterministic post-processing for exact dimensions and text. 3. Quote hosted work from a dated price source before submission; use local processing when adequate. 4. Generate a small, reviewable candidate set and compare each result with the approved references. 5. Repair bounded defects locally when possible; regenerate only when the defect is global. 6. Export to a new path and inspect dimensions, color mode, alpha, and compression artifacts. ## Quality gates - Exact copy and logos are composited from approved source files, not trusted to generative lettering. - Reference identity and product geometry are checked separately from aesthetic quality. - Upscaling, background removal, and color conversion are verified on edge cases. ## Deliverable Final image files, editable source where available, prompt/settings record, and a short quality report. Include the exact commands or application operations used, the files changed, and the checks that were actually run. ## Boundaries - Keep source files unchanged; write new versions to an explicit output path. - Do not transmit private assets or prompts to an external provider unless the user has explicitly approved that destination. - Do not run paid generation or destructive application operations without clear authorization and a stated budget or scope. - If a required application, plugin, model, or entitlement is absent, report the limitation and stop at the last verified stage. - Separate measured results from visual judgment, assumptions, and recommendations.
Versions
The database has no other skills named image-reference-edit.
Database snapshot: 13.11.2026 · package version 7442661ae8b8 (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 85image-editing
Edit supplied images while preserving product identity, exact copy, composition, and previously approved changes. Use when a supplied image needs reference-based edits or successive revisions. - 100p-image-edit
Use when someone wants to edit an existing photo — change outfits or backgrounds, compose from reference images, or apply prompt-driven edits. - 84edit-image
Edit, restyle, or vary an EXISTING image file using OpenAI gpt-image-2 through the Codex CLI — change one thing and keep the rest, swap a background, recolor to a brand palette, add or fix text, or produce style-matched variants of a reference image. Use whenever the user points… - 100codex-image-studio
Image generation workflow for Codex using OpenAI-compatible APIs. Use when the user wants Codex to create, edit, or batch-produce images through a third-party or self-hosted image endpoint, including text-to-image, image-to-image, multi-reference edits, mask edits, or multi-job generation. - 90image-edit
Edit or retouch an existing image with a connector - background removal, generative fill/expand, cleanup, recolor, or compositing. Use when modifying a photo or asset rather than generating from scratch. - 100fal-image-edit
Edit images using AI on fal.ai. Style transfer, object removal, background changes, and more. Use when the user requests "Edit image", "Remove object", "Change background", "Apply style", or similar image editing tasks.