Summary
Support sending images to the agent and displaying images in conversation — both user-uploaded images and any images returned by the agent in responses.
UX Requirements
- Paste images directly into chat input (Ctrl+V / Cmd+V)
- Image thumbnail preview in the input area before sending
- Inline image display in both assistant and user messages
- Click-to-expand lightbox for full-size image viewing
- Support for base64-encoded images and URL-referenced images
- Graceful fallback for unsupported image formats (show placeholder with format name)
- Responsive image sizing — fit message width, don't overflow
- Alt text support for accessibility
Implementation Notes
Images are sent as content blocks with type: image_url following the OpenAI multi-modal message format. Base64 encoding is handled client-side for paste and upload flows. URL-referenced images in assistant responses render directly via <img> tags. The lightbox component should be shared across the app (reusable for any future media display). Lazy-load images outside the viewport for long conversations.
Companion Issues
Companion issues filed on fips-agents/agent-template, fips-agents/gateway-template, fips-agents/fips-agents-cli, and fips-agents/examples.
Size
S-M
Summary
Support sending images to the agent and displaying images in conversation — both user-uploaded images and any images returned by the agent in responses.
UX Requirements
Implementation Notes
Images are sent as content blocks with
type: image_urlfollowing the OpenAI multi-modal message format. Base64 encoding is handled client-side for paste and upload flows. URL-referenced images in assistant responses render directly via<img>tags. The lightbox component should be shared across the app (reusable for any future media display). Lazy-load images outside the viewport for long conversations.Companion Issues
Companion issues filed on fips-agents/agent-template, fips-agents/gateway-template, fips-agents/fips-agents-cli, and fips-agents/examples.
Size
S-M