Crystallize Developer

Use this skill to execute Crystallize work with correct API selection, authentication, and production-safe integration patterns.

Quick Start

1. Identify the task type.
2. Select the correct API using references/api-map.md.
3. Apply the required auth model.
4. Implement query/mutation/webhook flow.
5. Validate behavior and production safety.

Task Routing

• Use Discovery API for storefront browse/search/filter/facet queries.
• Use Catalogue API for path-based deterministic reads.
• Use Shop API for cart and checkout operations.
• Use Core API for app/integration management operations (items, customers, orders, flows, price lists, etc.).
• Use Webhooks for event-driven downstream actions.
• Use Mass operations for large migrations/seeding/fixes with ordered bulk tasks.

Workflow

1. Confirm tenant and environment (dev/stage/prod).
2. Pick API + endpoint + auth from references/api-map.md.
3. Implement smallest working query/mutation first.
4. Add production constraints:

• Keep secrets server-side.
• Use idempotency for webhook consumers and bulk replays.
• Add retries with backoff for transient failures.
• Verify webhook signatures before processing.

5. Add observability:

• Log request IDs/correlation IDs and operation identifiers.
• Track webhook delivery failures and dead-letter retries.
• Track bulk task progress and operation logs for migrations.

Reference Files

• Read references/api-map.md for API selection, endpoints, auth headers, and security notes.
• Read references/workflows.md for execution patterns (storefront, checkout, migrations, webhook consumers).

Output Expectations

When delivering implementation output, include:

1. API choice and why.
2. Endpoint and auth headers/tokens required.
3. Minimal query/mutation/webhook handler code.
4. Security notes (token handling/signature verification).
5. Validation steps (local + staging).