Add a shared visiblePromptText helper and strip reference tokens,
image-generator directives, and model-selection lines from agent thread
summaries and the assistant panel header, so titles show the user's
actual prompt instead of internal directives. Covered by a server test.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Skip WebP re-encoding for PNG/AVIF/BMP uploads and any bitmap that
actually contains alpha, tag transparency-capable canvas uploads with
the transparent-image tone, and detect alpha in merged layers so the
composed node keeps a transparent tone.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Server: forward canvas/reference images to the image model so extract,
continue, and edit requests stay faithful to the source product.
- Add ImageRequestOptions.Images and route requests with inputs to the
/v1/images/edits endpoint, via multipart upload (default) or JSON
image_url payload, selectable through Agent.Image.InputImageTransport.
- Extract reference URLs from prompt directives, inline @image tokens,
and chat message image/mask contents; fold them into the idempotency
key and prepend a reference-fidelity instruction to the prompt.
Frontend:
- Shrink CanvasToast to a compact top pill.
- Paste an internal transparent clipboard node instead of re-uploading
when it matches the clipboard image.
- Keep agent-project polling alive on stream error while streaming.
root