feat(crisis): enable image-capable intake handler for test users #207

Open
Rockachopa wants to merge 1 commits from step35/201-intake-submission-from-image into main
Owner

Extend crisis.intake.handle_intake_submission to accept image data (OCR text, object labels, operator notes) and integrate with image_screening.screen_image_signals().

This addresses #201: "Intake submission from Image Tester" — the text-only intake handler existed (PR #202 / commit 77fc411) but lacked image support. Test users can now submit messages with attached image analysis data.

Changes:

  • Added image_ocr, image_labels, image_notes optional params
  • IntakeResult now includes image_signals and image_distress_score
  • Internal logging captures image distress in test metrics
  • Interactive CLI (python -m crisis.intake) supports [img:ocr=...] syntax

Backward compatible: text-only intake calls unchanged. All 8 intake tests pass.


Closes #201

Extend `crisis.intake.handle_intake_submission` to accept image data (OCR text, object labels, operator notes) and integrate with `image_screening.screen_image_signals()`. This addresses #201: "Intake submission from Image Tester" — the text-only intake handler existed (PR #202 / commit 77fc411) but lacked image support. Test users can now submit messages with attached image analysis data. Changes: - Added `image_ocr`, `image_labels`, `image_notes` optional params - IntakeResult now includes `image_signals` and `image_distress_score` - Internal logging captures image distress in test metrics - Interactive CLI (`python -m crisis.intake`) supports `[img:ocr=...]` syntax Backward compatible: text-only intake calls unchanged. All 8 intake tests pass. --- Closes #201
Rockachopa added 1 commit 2026-04-30 14:48:08 +00:00
feat(crisis): add image-capable intake handler for test users (closes #201)
All checks were successful
Sanity Checks / sanity-test (pull_request) Successful in 5s
Smoke Test / smoke (pull_request) Successful in 8s
30b9c0ceef
Extend crisis.intake.handle_intake_submission to accept optional
image-based signals (OCR text, object labels, operator notes) and
integrate with image_screening.screen_image_signals().

Changes:
- crisis/intake.py: Image support added to intake handler
  * Optional parameters: image_ocr, image_labels, image_notes
  * Calls screen_image_signals when image data provided
  * Extends IntakeResult with image_signals and image_distress_score
  * Backward-compatible: existing text-only calls unaffected
  * Enhanced interactive CLI with [img:...] inline syntax
- Updated _log_test_intake to record image signals in test metrics

This is the smallest concrete fix for #201: image-capable
intake submission. The text-only handler already existed via PR #202.

Tests: all 8 intake tests pass (existing behavior preserved).
Smoke: manual call verifies image signals detected.
All checks were successful
Sanity Checks / sanity-test (pull_request) Successful in 5s
Smoke Test / smoke (pull_request) Successful in 8s
This pull request doesn't have enough required approvals yet. 0 of 1 official approvals granted.
You are not authorized to merge this pull request.
View command line instructions

Checkout

From your project repository, check out a new branch and test the changes.
git fetch -u origin step35/201-intake-submission-from-image:step35/201-intake-submission-from-image
git checkout step35/201-intake-submission-from-image
Sign in to join this conversation.