mirror of
https://github.com/ruvnet/RuVector.git
synced 2026-05-22 19:56:25 +00:00
930fca916f
SSE Proxy Decoupling (ADR-130): - Fix ruvbrain-sse proxy: proper MCP handshake, session creation, drain polling - Fix internal queue endpoints: session_create keeps receiver, drain returns buffered messages - Add response_queues to AppState for SSE proxy communication - Skip sparsifier for >5M edge graphs (was crashing on 16M edges) - Add SSE_DISABLED/MAX_SSE env vars for configurable connection limits - Route SSE to dedicated mcp.pi.ruv.io subdomain (Cloudflare CNAME) - Serve SSE at root / path on proxy (no /sse needed) - Update all references from pi.ruv.io/sse to mcp.pi.ruv.io - Fix Dockerfile consciousness crate build (feature/version mismatches) Claude Code CLI Source Research (ADR-133): - 19 research documents analyzing Claude Code internals (3000+ lines) - Decompiler script + RVF corpus builder for all major versions - Binary RVF containers for v0.2, v1.0, v2.0, v2.1 (300-2068 vectors each) - Call graphs, class hierarchies, state machines from minified source Integration Strategy (ADR-134): - 6-tier integration plan: WASM MCP, agents, hooks, cache, SDK, plugin - Integration guide with architecture diagrams and performance targets Co-Authored-By: claude-flow <ruv@ruv.net> |
||
|---|---|---|
| .. | ||
| .svelte-kit | ||
| docs | ||
| scripts | ||
| src | ||
| static | ||
| tests | ||
| cloud-run.yaml | ||
| Dockerfile | ||
| dragnes.config.ts | ||
| package.json | ||
| postcss.config.js | ||
| README.md | ||
| svelte.config.js | ||
| tailwind.config.cjs | ||
| tsconfig.json | ||
| vite.config.ts | ||
DrAgnes -- Dermatology Intelligence Platform
AI-powered dermoscopy analysis with collective learning, built on the RuVector embedding engine and pi.ruv.io brain network.
Quick Start
cd examples/dragnes
npm install
npm run dev
The app starts at http://localhost:5173.
Architecture
DrAgnes is a standalone SvelteKit application that provides:
- On-device CNN inference via MobileNetV3-Small (WASM) for dermoscopy image classification across 7 HAM10000 lesion classes (akiec, bcc, bkl, df, mel, nv, vasc).
- ABCDE scoring with asymmetry, border, color, diameter, and evolution metrics.
- Grad-CAM attention overlays showing which regions influenced the classification.
- HAM10000 demographic adjustment using Bayesian priors from clinical data (age, sex, body location).
- Privacy-first design with differential privacy (epsilon=1.0), EXIF stripping, and local-only processing by default.
- Brain sync via pi.ruv.io for federated learning and collective case sharing (opt-in).
- Offline PWA with service worker caching for model weights and static assets.
Directory Structure
src/
lib/
dragnes/ # Core library: classifier, types, preprocessing, privacy
components/ # Svelte UI components: capture, results, charts, panel
routes/
+page.svelte # Main page (loads DrAgnesPanel)
api/ # Server endpoints: health, analyze, feedback, similar
static/ # PWA manifest, icons, service worker
tests/ # Unit and benchmark tests
scripts/ # Deployment and analysis scripts
docs/ # Research documentation
API Endpoints
| Method | Path | Description |
|---|---|---|
| GET | /api/health |
Health check with model info |
| POST | /api/analyze |
Classify an image embedding |
| POST | /api/feedback |
Submit clinician feedback |
| GET | /api/similar/[id] |
Find similar cases |
Testing
npm run test
Deployment
# Build for production
npm run build
# Deploy to Cloud Run
npm run deploy
See cloud-run.yaml for the Cloud Run service configuration and Dockerfile
for the container build.