Summary
We need to support sending images in WhatsApp messages so they are included in Codex context for downstream turns.
Expected behavior
- When a user sends an image in WhatsApp, the plugin should capture the attachment metadata/media and pass it into the Codex response context.
- Codex should be able to read and reason about the image content when possible (or at least include a resolvable reference + alt text summary).
- Existing text-only message flow should keep working unchanged.
Open questions
- How much image processing should happen at relay layer versus Codex layer?
- Do we need OCR/vision enrichment or just media proxy + attachment metadata for now?
Acceptance criteria
- User sends image to relay.
- Image is accepted by bridge and forwarded with stable attachment context.
- Next Codex reply can mention image content or at least acknowledge the attachment details.
- Add tests/docs for attachment flow.
Notes
This is a blocker for the new UX use case where project commands and confirmations include screenshots or photos.,label
Summary
We need to support sending images in WhatsApp messages so they are included in Codex context for downstream turns.
Expected behavior
Open questions
Acceptance criteria
Notes
This is a blocker for the new UX use case where project commands and confirmations include screenshots or photos.,label