Focus: Review extraction implementations for reliability, idempotency, and operational safety. Verify that connectors handle schema drift, network failures, and partial extractions without data loss or duplication.

Produces: Review findings for each extraction job covering idempotency, error handling, schema drift resilience, and operational readiness.

Reads: Extractor's implementation, source catalog from discovery.

Anti-patterns:

• Approving extraction logic without verifying idempotency (re-run safety)
• Not testing what happens when a source schema changes mid-extraction
• Ignoring partial failure scenarios (e.g., network timeout after 80% of records)
• Treating retry logic as optional for "reliable" sources
• Not verifying that extraction metadata is sufficient for debugging production issues

Focus: Implement extraction logic that reliably moves data from sources to the staging area. Handle incremental loads, rate limiting, error recovery, and extraction metadata tracking. Prioritize correctness and idempotency over speed.

Produces: Extraction jobs for each source with full-load and incremental-load paths, error handling, retry logic, and extraction metadata (batch ID, timestamp, source identifier).

Reads: Source catalog and schema analysis from discovery, source system API documentation.

Anti-patterns:

• Building only full-load extraction when incremental is feasible
• Ignoring source system rate limits or connection pool constraints
• Silently dropping records on extraction errors instead of dead-lettering
• Not tracking extraction metadata (when, what, how much) for auditability
• Hardcoding connection strings or credentials instead of using config/secrets management