What it does
Instant and needs no calibration session: it reads the slider adjustments the user has already made in imagic and feeds them into the same style profile that calibration builds. By default it learns from every edited photo in the library, most recently edited first, up to limit; pass photo_ids to train on a specific set instead. Safe to re-run any time: each run refines the profile with the photos' current edits. Use it after start_style_calibration to deepen a fresh profile, or on its own to train from a back catalog of edited photos with no session at all. Returns learned and skipped counts, the scenarios covered, and a profile summary.
Parameters
| Name | Type | Required |
|---|---|---|
photo_ids |
list[int] | None |
optional |
limit |
int |
optional |
Asking for it
You do not call this by name. You say what you want, and the agent picks the tool:
Learn my style from the photos I have already edited, then apply it to the rest of this shoot.
The rest of the style tools
Workflows that use it
- Teach an AI your editing style — AI photo editing in your own style
Where it runs
On your machine. imagic's MCP server is a local process that your AI client starts; the photos never leave your disk and nothing is uploaded. That is true of every tool in the reference, including this one.
To call it you need an MCP client connected to imagic. Setup takes about three minutes for Claude Code, Cursor, VS Code, Codex, Gemini CLI and ten others.
Give your AI the whole pipeline
learn_style_from_library is one of 18 tools in imagic Max. The free 7-day trial includes all of them.