Scaffolds and validates ODCS v3.1.0 data contracts using datacontract-cli. Use when the user asks to create, write, scaffold, or author a data contract, generate an ODCS file, define a contract for a table or model, or mentions ODCS, datacontract-cli, logicalType, dimension, data product contract, or schema contract. Also trigger for: "new contract", "add contract", "contract for a model", "lint contract", or questions about valid logicalType and dimension enum values. Do not use for general dbt model schema documentation unrelated to contracts, SQL quality checks outside of a contract context, or PySpark/Spark pipeline code.
Scaffolds new dbt models (staging, dimension, or fact) following dbt best practices. Use when the user asks to create a dbt model, scaffold a staging model, add a mart model, build a dim_ or fct_ table, generate a schema.yml entry, or add a sources.yml source. Also trigger for: dbt staging, stg_ prefix, dim_ prefix, fct_ prefix, surrogate key, dbt_utils.generate_surrogate_key, schema.yml, sources.yml, dbt ref(), dbt source(), mart model, dimension table, fact table, dbt scaffold, new dbt model. Do NOT use for reviewing existing dbt models, debugging dbt run failures, or generating dbt documentation outside of schema.yml entries.
Produces production-quality PySpark code, unit tests, and project layouts for Spark 3.5+ with Delta Lake. Use when the user asks to write, review, refactor, or debug PySpark code, define schemas, design ETL modules, or discuss DataFrame API best practices. Also trigger for: PySpark, Spark DataFrame, SparkSession, pyspark.sql.functions, Delta Lake, MERGE INTO, time travel, OPTIMIZE, ZORDER, VACUUM, Change Data Feed, readStream, writeStream, Structured Streaming, foreachBatch, or assertDataFrameEqual. Do NOT use for generic Python questions, non-Spark data tools (pandas, Polars, DuckDB), or writing pytest tests (use pytest-data-engineering for that).
Write pytest tests for data engineering projects in Python. Trigger when the user asks to write, add, or refactor unit tests for data engineering code โ PySpark transformations, pandas pipelines, DuckDB queries, Polars pipelines, ETL modules, or data quality checks. Also trigger on pytest, conftest, fixtures, _test.py, _data.py, assertDataFrameEqual, pytest.mark, coverage, or testing-strategy requests. Trigger phrases: "write a test", "add tests", "test this transformation", "how do I test this", or "set up pytest". Also trigger when the user asks to set up a virtualenv, configure ruff, set up pre-commit hooks, or configure a CI pipeline for a data project. Do NOT use for writing production PySpark code (use pyspark-expert), general Python testing outside data engineering, or tasks that do not end in a pytest test file.
Engine-agnostic advisor for the Write-Audit-Publish (WAP) data-quality pattern โ stage an isolated write, audit it, and atomically publish only clean data so consumers never see bad data. Use when designing or reviewing a pipeline that needs a quality gate before exposing results, or when someone mentions WAP, write-audit-publish, staging-then-publish, audit gate, atomic swap/publish, blue-green data, Iceberg WAP branch/cherry-pick, "don't publish bad data", "validate before exposing", or quality SLAs on a published dataset. Advises on design and mechanism choice; does not generate production pipeline code. Not for generic dbt test writing unrelated to a publish gate, dimensional modeling, or ingestion/CDC design.
Produces a numbered Architecture Decision Record (ADR) file in the project's ADR directory. Use when the user asks to create an ADR, write an architecture decision record, document a decision, add to the decision log, or mentions docs/adr/. Also trigger when the user says "new ADR", "log a decision", or "record why we chose X". Do not use for general documentation, READMEs, or non-decision narrative writing.
Creates a pull request for the current branch against the base branch by inspecting recent commits and changed files to infer the PR title, type, and description. Use when the user asks to open a PR, create a pull request, push and open a PR, submit a PR, or mentions git push with PR creation. Also trigger when the user says "create PR", "open a pull request", "fill the PR template", or "write a PR description". Do not use for committing code, rebasing, code review, or branch management tasks that do not end in creating a PR.
Applies a disciplined research method to conceptual, definitional, and "difference between X and Y" questions โ especially overloaded terms and LLM/agent-domain topics, where answering from memory is a known trap. Use when asked to explain or compare concepts, define a term spanning multiple communities/lineages (e.g. "sensor", "hook", "agent"), or when the user demands primary sources. Also trigger when a question names Claude, Anthropic, or any agent-engineering concept with the provider unstated. Surfaces competing definitions first, checks the local repo/config before the literature, fetches and quotes primary sources instead of trusting search snippets, and says explicitly when the canon is unsettled instead of forcing false symmetry. Not for Claude API/model reference lookups (pricing, params, context windows, model IDs) โ use claude-api for those. Not for code-writing, and not a generic web-search wrapper; skip when a single canonical source is already known and correctly cited.