Images
Edit an image
Edit an existing image using a natural-language instruction.
Use this skill in a chatOpens a new chat with it attached.
What it asks you for
- Imagerequired
- The image to edit.
- What to changerequired
- Describe the edit.
- Image modelrequired
- Which model generates it.
What it tells the AI to do
The full instructions, exactly as they run.
Edit this image.
- Source: {{ image }}
- Instruction: {{ instruction }}
- Model: {{ model }}
Use `inference image` with the source as a reference image.
Change ONLY what was asked. An "edit" that silently restyles the whole frame is
a new image, not an edit — if the instruction cannot be done without changing the
rest, say so and offer both.
If the user marked a region (a `<marker>` with a `box`), that is the edit area:
use its coordinates rather than re-interpreting the instruction spatially.
## Delivery
Write the finished artifact into Drive. Never upload a user's content to a
third-party file host (`0x0.st`, `catbox.moe`, `transfer.sh`, pastebins, any
off-idapt URL) — on a computer, render to `/tmp/output/` and `computer file-download`
it into Drive. If you catch yourself reaching for `curl -F`, stop: you already
hold the artifact.- image
- edit
More images skills
Run it on your own work
idapt gives this skill your files, 200+ models, and a real computer to work on. Start a chat and it is already attached.
Use Edit an image