メニュー
Drives data implementation with contracts, assertions, and validation evidence. Use when adding or changing ingestion logic, transformations, schemas, or published data products.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Forces replay-safe rollout plans, reconciliation gates, and rollback paths before executing any data backfill or historical reprocessing. Use when running /backfill, rerunning pipelines, repairing publish windows, or any work that risks double-counting or downstream corruption.
Helps agents classify data engineering work, choose the right preset and skill bundle, and pick the safest next command. Use when starting a session, triaging an ambiguous request, or deciding how to proceed.
Guides agents through resiliency testing for data platforms. Use when designing or running failure drills, recovery validation, failover tests, replay-safety checks, dependency outage exercises, or fault injection for pipelines and publishes.
Guides agents through production data incidents. Use when a pipeline fails, publishes bad data, misses an SLA, partially loads, corrupts state, or requires rollback, replay, or stakeholder communication.
Enforces production Kafka guardrails including non-breaking schema evolution, dead-letter queues for poison messages, and acks=all producer durability. Use when designing or changing Kafka topics, producers, consumers, schema registry policies, or streaming recovery paths.
Guides agents to wire Model Context Protocol servers for live data platform observability including Spark execution plans, OOM diagnosis, Kafka consumer lag, and orchestration run state. Use when agents need cluster metrics, streaming lag, or job diagnostics instead of blind code changes.
| name | data-quality-and-contract-testing |
| description | Drives data implementation with contracts, assertions, and validation evidence. Use when adding or changing ingestion logic, transformations, schemas, or published data products. |
Data work is not complete when code runs. It is complete when source assumptions, output contracts, and quality checks prove the behavior is correct.
Do not use this only as a final cleanup step. It should guide implementation from the start.
Define the contract before implementation. Capture:
Write the validation plan first. Common validations:
Reproduce data bugs with a failing check. If an incident or defect exists, write the failing validation or test before changing the pipeline.
Implement the smallest change that satisfies the contract.
Run the relevant validations and capture evidence. Evidence may include:
| Rationalization | Reality |
|---|---|
| "The warehouse query looks right." | Visual inspection does not scale and misses edge cases. |
| "We will add checks after the model stabilizes." | Unchecked pipelines create low-trust data and harder incident response. |
| "A successful job means the data is valid." | Jobs succeed while still producing incorrect or incomplete data. |
