Quantum Agent Engineering

QuantinuumClaw – Quantum Guppy/Selene Stack

This skill provides everything needed to build production-ready quantum applications using Quantinuum (hardware/emulator), Guppy (quantum language), Selene (FastAPI backend), and Fly.io (deployment), with optional Lovable frontend. It is tuned for the OpenClaw Clinical Hackathon and general quantum web apps.

When to Use This Skill

Use when:

• Building for the OpenClaw Clinical Hackathon or any clinical/healthcare quantum project
• Building web applications that use quantum computing (optimization, chemistry, ML, random, crypto)
• Deploying quantum algorithms as REST APIs or creating dashboards for quantum results
• User mentions: clinical, healthcare, drug discovery, treatment optimization, patient stratification, molecular simulation, clinical trials, Guppy, Selene, Fly.io

Example requests: "Build a quantum portfolio optimizer with a web interface" · "Deploy my Guppy algorithm to the cloud" · "Create a clinical molecular simulation demo" · "Set up a quantum ML service on Fly.io"

Stack at a Glance

Quick Start (One Command)

From the repo root:

python3 scripts/create_quantum_app.py \
  --app-name "clinical-demo" \
  --use-case "chemistry" \
  --description "Clinical molecular simulation" \
  --deploy

Then set VITE_API_URL in the frontend to your Fly.io app URL (e.g. https://clinical-demo.fly.dev).

Clinical use-case → --use-case mapping:

General use cases (portfolio, finance, etc.) also use optimization, chemistry, ml, random, crypto, or finance. See references/clinical-use-cases.md for detailed clinical mappings.

Full Workflow: Creating a Quantum Application

Step 1: Define the use case

Identify the problem (optimization, simulation, ML, cryptography, clinical, etc.).

Step 2: Create Selene backend

python3 scripts/setup_selene_service.py \
  --app-name "my-quantum-app" \
  --use-case "chemistry" \
  --description "Quantum chemistry simulator"

This creates a backend dir with FastAPI, health check, Dockerfile, and fly.toml.

Step 3: Implement your Guppy circuit

Edit my-quantum-app/main.py → QuantumService._run_real_quantum(). Use references/guppy_guide.md for syntax. For clinical: chemistry (molecule, shots, precision), optimization (objective, constraints), ML (features, epochs).

Step 4: Deploy to Fly.io

python3 scripts/flyio_deploy.py --app-name "my-quantum-app" --service-dir "my-quantum-app" --region "lhr"

Set secrets with fly secrets set; use emulator for demos if preferred.

Step 5: Frontend

Use assets/lovable-template/ or run:

python3 scripts/lovable_integrate.py \
  --app-name "my-frontend" \
  --backend-url "https://my-quantum-app.fly.dev" \
  --quantum-use-case "chemistry"

Then npm install and npm run dev in the frontend dir.

Step 6: Connect and test

Point frontend VITE_API_URL to the Fly.io backend; hit /health to verify.

Clinical Use Case Cheat Sheet

• Drug discovery / molecular simulation: chemistry — VQE-style energy/property in Guppy; expose molecule type and params via API.
• Treatment / resource optimization: optimization — Define objective (cost, wait time); run QAOA in Selene; display results in UI.
• Patient stratification / classification: ml — Map patient features to model inputs; return risk/stratum or classification.
• Randomization (e.g. trials): random — Quantum RNG from Guppy; expose bits/shots in API.
• Security / key material: crypto — Key generation or quantum-safe primitives; keep keys on backend only.

Data and Compliance (Clinical / Hackathon)

• Demos: Use synthetic or de-identified data only. Do not send real PHI to quantum backends or store in Fly.io without a compliance plan.
• API keys: Store in Fly.io secrets (fly secrets set), never in code or frontend.
• Production: Add auth, rate limiting, and consider HIPAA/DPA; restrict CORS in Selene.

Resources

scripts/

• create_quantum_app.py — All-in-one: backend + deploy + frontend
• setup_selene_service.py — Scaffold Selene backend
• flyio_deploy.py — Deploy to Fly.io
• lovable_integrate.py — Frontend wired to backend URL

references/

• guppy_guide.md — Guppy syntax, gates, circuits, examples
• selene_api.md — Endpoints, request/response, errors, jobs
• flyio_config.md — Fly.io scaling, regions, secrets, monitoring
• lovable_patterns.md — Frontend patterns, dashboard, API client
• clinical-use-cases.md — Detailed clinical use-case mappings and compliance notes

assets/

• selene-template/ — Backend boilerplate (main.py, Dockerfile, fly.toml, .env.example)
• lovable-template/ — React/TS frontend with QuantumDashboard and API client

Advanced: Multi-Quantum Use Cases

• Optimization dashboard: Selene + QAOA/VQE; Lovable with sliders; Fly.io with scaling.
• Chemistry explorer: Guppy molecular simulations; 3D viewer; optional persistent storage.
• Quantum ML API: Selene exposing QNN/QSVM; Lovable for training/predictions; Fly.io (GPU if needed).

Performance, Cost, and Security

• Queuing: Use job queues for long-running quantum jobs; WebSockets or polling in frontend.
• Caching: Cache identical computations to reduce quantum hardware cost.
• Fly.io: Scale to zero when idle (min_machines_running = 0); use references/flyio_config.md for VM sizing.
• Security: No API keys in frontend; rate limiting on Selene; HTTPS in production; see references/selene_api.md for auth patterns.

Troubleshooting

• Guppy import error: pip install guppy in backend; or use mock mode for demos.
• Selene not starting: Check fly.toml, fly logs, and env vars.
• Frontend can’t connect: Verify VITE_API_URL, CORS in Selene, and curl .../health.
• Fly.io deploy fails: fly deploy --clean; fly logs --phase build; ensure fly auth login.
• Quantum results wrong: Validate circuit logic and measurement; test with emulator first.

Next Steps

After initial setup: monitor quantum usage/costs; add auth to Selene if public; improve error handling and logging; consider persistence for job history.

---

For detailed clinical use-case specs and compliance reminders, see references/clinical-use-cases.md.