pediatric-ai-scribe-v3/public
Daniel 07d1c541a0
Some checks failed
Forgejo Android APK / Root app tests (push) Successful in 50s
Forgejo Docker Build / Root app tests (push) Successful in 48s
Forgejo Android APK / Build signed APK (push) Successful in 2m2s
Forgejo Docker Build / Build Docker image (push) Successful in 23s
Forgejo Docker Build / Deploy to the host (push) Failing after 0s
fix: a deck the model fumbles once is asked for again, not abandoned
A presentation is generated as a typed deck — the model chooses each slide's
layout — and that deck is what scripts/render_pptx.py renders. When the reply
did not parse as a deck, the route fell straight back to asking for markdown,
and slideSpec.build() then inferred slides from it. Both paths go through
python-pptx, but the fallback's layout is guessed from heading and list
structure rather than chosen, so everything lands as title-and-bullets.

Measured on the stored library: since decks landed, 7 of 8 generations produced
one and 1 did not. Models are stochastic, so one unlucky reply was costing the
whole layout. It now asks a second time with the same prompt before giving up.

The fallback was also invisible. It warned to the console, where the person who
would simply have generated again could not see it, so they kept the plainer
deck without knowing a better one was one click away. The response now carries
deckFallback and the UI says it came out as plain slides, and why.

Fixed the reason heuristic while adding tests for it: truncation was claimed for
any reply not ending in "}", which is every prose refusal. It is now only
claimed for a reply that began as JSON and stopped.

The four generate tests run the handler. Verified against a mutation: removing
the retry fails tests 6 and 7 and nothing else.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Dv6sqaY6Vq3ChZHMem3cnU
2026-09-12 05:34:14 +02:00
..
.well-known Add SHA256 fingerprint to assetlinks.json for TWA domain verification 2026-03-29 00:32:58 +00:00
audio Match TTS voices to the selected LiteLLM model 2026-07-30 17:34:34 +02:00
components fix: /api/health/detailed reported a text-to-speech provider that does not exist 2026-09-12 05:05:39 +02:00
css refactor: cut the explanatory copy from the sign-in screen and My Resources 2026-09-11 17:57:07 +02:00
data move pe guide data to json 2026-05-08 08:56:21 +02:00
icons Feat: add stethoscope+pencil favicon in SVG, ICO, and PNG sizes 2026-03-22 14:09:18 -04:00
images/pe-guide feat(pe-guide): real audio, innocent-murmur panel, APTM image, all-ages resp+CV 2026-04-22 21:00:42 +02:00
img feat: Bedside clinical reference module + age→weight estimator + dose-math unit tests 2026-04-20 02:49:42 +02:00
js fix: a deck the model fumbles once is asked for again, not abandoned 2026-09-12 05:34:14 +02:00
vendor fix: recordings that produced nothing, and the boxes that zoomed on iOS 2026-09-11 00:16:19 +02:00
404.html rebrand to Pediatric Clinical Tools, add diagrams tab, simplify litellm transcribe/tts 2026-05-02 18:23:06 +02:00
e2e-harness.html test(e2e): repair the harness, taking the browser suite from 96 failures to 11 2026-09-11 18:54:03 +02:00
favicon.ico Feat: add stethoscope+pencil favicon in SVG, ICO, and PNG sizes 2026-03-22 14:09:18 -04:00
index.html feat: sign in with a code emailed to you, offered beside the password 2026-09-11 20:12:03 +02:00
manifest.json fix: image models can be added and offered; solid phone top bar on iOS 2026-09-10 13:46:19 +02:00
sw.js harden clinical assistant source handling 2026-05-09 19:59:04 +02:00
template-guide.md feat: add extension import preview 2026-05-08 22:26:53 +02:00