stackchain/timmy-talking-turd
2
0 Fork 0
Code Issues 21 Pull Requests 7 Actions 4 Packages Projects Releases 41 Wiki Activity

Field finding: hard-stool photo should trigger soreness, hydration, and tarry-color follow-up #67

New Issue
Closed
opened 2026-08-27 00:02:15 +00:00 by rockachopa · 1 comment
rockachopa commented 2026-08-27 00:02:15 +00:00
Member
Copy link
Copy source

Field finding (anecdotal; not labeled clinical ground truth)

A real photo-first conversation exposed a useful follow-up pattern:

  • Visible stool appeared small, hard/lumpy, and dark brown, provisionally consistent with Bristol 1–2.
  • The user then self-reported mild anal soreness and mild dehydration.
  • The crucial color clarification was whether the stool was merely dark brown versus genuinely black, sticky, and tar-like.

Do not retain the submitted image or treat this observation as classifier training truth. Pain, hydration, texture/ease of passage, and other symptoms remain user-reported—not inferred from pixels.

Product implication

After a user confirms a hard/lumpy photo suggestion, Timmy should offer a short, non-diagnostic follow-up that captures:

  1. Was it hard or difficult to pass?
  2. Any soreness or sharp pain afterward?
  3. Any visible red blood?
  4. Was it dark brown, or truly black/tarry/sticky?
  5. Does the user feel dehydrated / is urine unusually dark?

The response can give proportionate self-care guidance for an otherwise mild report (steady hydration, avoid straining, gentle cleaning, warm sitz bath, external barrier ointment, food-based fiber) while preserving the authoritative urgent red-flag boundary.

Acceptance criteria

  • Photo analysis continues to suggest only visible form/color/quality.
  • Soreness, dehydration, difficulty passing, and tarry/sticky character require explicit user confirmation.
  • Dark brown is not silently escalated as black stool.
  • Confirmed black/tarry stool, substantial blood, severe/worsening pain, dizziness/weakness, vomiting, or very low urine triggers the existing deterministic safety path.
  • Mild guidance is concise, non-diagnostic, and does not imply that a photo ruled out bleeding or disease.
  • Add conversation tests for the hard-stool → soreness/dehydration follow-up and dark-brown-vs-tarry clarification.

Related: #5, #23, #62.

## Field finding (anecdotal; not labeled clinical ground truth) A real photo-first conversation exposed a useful follow-up pattern: - Visible stool appeared small, hard/lumpy, and dark brown, provisionally consistent with Bristol 1–2. - The user then self-reported mild anal soreness and mild dehydration. - The crucial color clarification was whether the stool was merely dark brown versus genuinely black, sticky, and tar-like. Do **not** retain the submitted image or treat this observation as classifier training truth. Pain, hydration, texture/ease of passage, and other symptoms remain user-reported—not inferred from pixels. ## Product implication After a user confirms a hard/lumpy photo suggestion, Timmy should offer a short, non-diagnostic follow-up that captures: 1. Was it hard or difficult to pass? 2. Any soreness or sharp pain afterward? 3. Any visible red blood? 4. Was it dark brown, or truly black/tarry/sticky? 5. Does the user feel dehydrated / is urine unusually dark? The response can give proportionate self-care guidance for an otherwise mild report (steady hydration, avoid straining, gentle cleaning, warm sitz bath, external barrier ointment, food-based fiber) while preserving the authoritative urgent red-flag boundary. ## Acceptance criteria - [ ] Photo analysis continues to suggest only visible form/color/quality. - [ ] Soreness, dehydration, difficulty passing, and tarry/sticky character require explicit user confirmation. - [ ] Dark brown is not silently escalated as black stool. - [ ] Confirmed black/tarry stool, substantial blood, severe/worsening pain, dizziness/weakness, vomiting, or very low urine triggers the existing deterministic safety path. - [ ] Mild guidance is concise, non-diagnostic, and does not imply that a photo ruled out bleeding or disease. - [ ] Add conversation tests for the hard-stool → soreness/dehydration follow-up and dark-brown-vs-tarry clarification. Related: #5, #23, #62.
rockachopa added the type/featurepriority/P0area/safety labels 2026-10-09 08:11:07 +00:00
timmy was assigned by rockachopa 2026-10-09 08:11:07 +00:00
rockachopa added this to the M1 — Sovereign Photo Intelligence milestone 2026-10-09 08:11:07 +00:00
rockachopa commented 2026-10-09 08:11:19 +00:00
Author
Member
Copy link
Copy source

Triaged and implemented in this pulse.

  • Triage: labels added (area/safety, priority/P0, type/feature), milestone set to M1, assigned to timmy.
  • Delivery: #70 (branch timmy/67-hard-stool-followup at 4cc5f4f, not merged — awaiting review).

The hard/lumpy (Type 1-2) follow-up asks all five user-reported questions with nothing pre-answered, gives non-diagnostic self-care for a mild report (explicitly stating the photo cannot rule out bleeding), never escalates dark brown, and routes confirmed visible blood or truly black/tarry stool into the same deterministic red-flag path manual logging uses.

Evidence on the branch: npm test 104/104, npm run test:followup PASS in a real browser, all six existing acceptance suites still PASS, syntax checks PASS, and scripts/record_release_demo.mjs records the follow-up, the mild guidance, and the black/tarry alert handoff (71.8s, synthetic data only).

Deliberately deferred: persisting the mild answers (difficulty passing, soreness, hydration) as ledger fields — that is a schema change owned by #35/#61, not a safety PR. Say the word if you want it filed as its own ticket.

Triaged and implemented in this pulse. - Triage: labels added (`area/safety`, `priority/P0`, `type/feature`), milestone set to M1, assigned to `timmy`. - Delivery: https://forge.alexanderwhitestone.com/git/stackchain/timmy-talking-turd/pulls/70 (branch `timmy/67-hard-stool-followup` at 4cc5f4f, not merged — awaiting review). The hard/lumpy (Type 1-2) follow-up asks all five user-reported questions with nothing pre-answered, gives non-diagnostic self-care for a mild report (explicitly stating the photo cannot rule out bleeding), never escalates dark brown, and routes confirmed visible blood or truly black/tarry stool into the same deterministic red-flag path manual logging uses. Evidence on the branch: `npm test` 104/104, `npm run test:followup` PASS in a real browser, all six existing acceptance suites still PASS, syntax checks PASS, and `scripts/record_release_demo.mjs` records the follow-up, the mild guidance, and the black/tarry alert handoff (71.8s, synthetic data only). Deliberately deferred: persisting the mild answers (difficulty passing, soreness, hydration) as ledger fields — that is a schema change owned by #35/#61, not a safety PR. Say the word if you want it filed as its own ticket.
Sign in to join this conversation.
No Branch/Tag Specified
Labels
Clear labels
area/backend
Application API and storage
 area/data
Dataset and provenance
 area/frontend
PWA and interaction design
 area/model
Vision model and inference
 area/privacy
Privacy and consent
 area/release
Release engineering and distribution
 area/safety
Clinical safety boundary
 blocked
Blocked on an explicit dependency or decision
 human-gate
Requires explicit human approval
 priority/P0
Release-critical
 priority/P1
High priority
 priority/P2
Planned
 state:review
type/docs
Documentation and evidence
 type/epic
Roadmap epic
 type/feature
Product feature
 type/ops
Infrastructure and operations
 type/research
Research or evaluation
 type/security
Security hardening
 type/test
Verification and quality gate
No labels area/safety priority/P0 type/feature
Milestone
No items
No Milestone M1 — Sovereign Photo Intelligence
Projects
Clear projects
No projects
Assignees
Clear assignees
timmy (Timmy · the Wizard Mind) zephyrus-sparkborn (Zephyrus Sparkborn)
No Assignees timmy
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: stackchain/timmy-talking-turd#67