| Dimension | Shared | Only in A | Only in B | Overlap |
|---|---|---|---|---|
| Sections | 0 | 1 | 1 | 0% |
| Commands | 0 | 0 | 0 | — |
| Section tags | 0 | 2 | 1 | 0% |
What each file covers
Sections
0 shared · 1 only in A · 1 only in B- − ML Training Memory
- + Supabase Rules
Commands
neither file has anySection tags
0 shared · 2 only in A · 1 only in B- − code-style
- − performance
- + do-not
Line diff
Aledon8/OpenLeukemia · ml/training/CLAUDE.md
@@ −1 @@
1# ML Training Memory
2
3Apply this inside `ml/training/`.
4
5- Prefer reproducible training scripts with explicit inputs, seeds, outputs, and metrics.
6- Keep patient-identifying data out of the repository.
7- Document label assumptions and cohort limitations near the training code.
8- Structured models support signals and trends, not diagnosis.
9
Aledon8/OpenLeukemia · .cursor/rules/supabase.mdc
@@ +1 @@
1---
2description: Supabase data, storage, and RLS rules
3globs:
4 - "supabase/**/*"
5alwaysApply: false
6---
7
8# Supabase Rules
9
10- Use Supabase for auth, CRUD, storage, consent records, RLS, and lightweight Edge Functions.
11- Keep heavy ML, OCR, extraction, parsing, and model workflows in `ai-service`.
12- Treat RLS as a primary protection layer.
13- Pair sensitive tables with narrow, auditable policies.
14- Plan affected data flows and access paths before schema or policy changes.
15
@@ −1 +1 @@
1−# ML Training Memory
1+---
2+description: Supabase data, storage, and RLS rules
3+globs:
4+ - "supabase/**/*"
5+alwaysApply: false
6+---
27
3−Apply this inside `ml/training/`.
8+# Supabase Rules
49
5−- Prefer reproducible training scripts with explicit inputs, seeds, outputs, and metrics.
6−- Keep patient-identifying data out of the repository.
7−- Document label assumptions and cohort limitations near the training code.
8−- Structured models support signals and trends, not diagnosis.
10+- Use Supabase for auth, CRUD, storage, consent records, RLS, and lightweight Edge Functions.
11+- Keep heavy ML, OCR, extraction, parsing, and model workflows in `ai-service`.
12+- Treat RLS as a primary protection layer.
13+- Pair sensitive tables with narrow, auditable policies.
14+- Plan affected data flows and access paths before schema or policy changes.
915
