Data Airflow DBT Orchestration
Use when the user asks to run dbt from Airflow, Astronomer Cosmos, DbtCloudRunJobOperator, dbt in an Airflow 3 DAG, Asset-based scheduling or orchestrate a daily ETL with Airflow and dbt. Produces the integration pattern, DAG code, model-level retries and CI. Not for general ETL architecture (use data-etl-pipeline).
Install
npx skills add j4flmao/agent-skills --skill data-airflow-dbt-orchestrationgit clone --depth 1 https://github.com/j4flmao/agent-skills.git /tmp/agent-skills
mkdir -p ~/.claude/skills
cp -r /tmp/agent-skills/skills/data/etl-pipelines ~/.claude/skills/data-airflow-dbt-orchestrationPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Airflow + dbt Orchestration
## Purpose
Run dbt transformations on a schedule or when upstream data lands, with retries, alerting and lineage, without turning Airflow into a compute engine. Airflow decides *when* and *in what order*; the warehouse does the work; dbt defines the SQL models and tests. The main design decision is how much of the dbt graph Airflow should see.
## Agent Protocol
### Trigger
"airflow dbt", "run dbt from airflow", "cosmos", "DbtCloudRunJobOperator", "dbt task group", "dbt in dag", "airflow 3 migration dbt", "orchestrate dbt".
### Input Context
- Airflow version/distribution (OSS, Astronomer, MWAA, Cloud Composer), dbt Core vs dbt Cloud (platform), warehouse.
- Project size (number of models), runtime, SLAs, upstream ingestion triggers.
### Output Artifact
Integration choice + DAG code + dbt selectors/tests + CI steps. No file unless requested.
### Response Format
```
Pattern: {Cosmos per-model|dbt Cloud job|single dbt build task|KubernetesPodOperator}
Schedule: {cron|Asset-triggered}
Selectors: {dbt --select ...}
Failure handling: {retries, alerts, partial reruns}
```
### Completion Criteria
- [ ] DAG uses Airflow 3 APIs (`schedule=`, `airflow.sdk` imports), `catchup` set explicitly.
- [ ] dbt tests run in the same flow (`dbt build`) and failures stop downstream publishing.
- [ ] Credentials come from Airflow connections/secret backend, never from code.
- [ ] Reruns possible for a failed model subset.
## Integration PatternsDetails
- Skill name
data-airflow-dbt-orchestration- Repository
- j4flmao/agent-skills
- Path
skills/data/etl-pipelines/SKILL.md- License
- MIT
- Repo updated
- 17 November 2026
FAQ
What does the Data Airflow DBT Orchestration skill do?
Use when the user asks to run dbt from Airflow, Astronomer Cosmos, DbtCloudRunJobOperator, dbt in an Airflow 3 DAG, Asset-based scheduling or orchestrate a daily ETL with Airflow and dbt. Produces the integration pattern, DAG code, model-level retries and CI. Not for general ETL architecture (use data-etl-pipeline).
How do I install the Data Airflow DBT Orchestration skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add j4flmao/agent-skills --skill data-airflow-dbt-orchestration. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Data Airflow DBT Orchestration popular?
It lives in j4flmao/agent-skills, which has 45 GitHub stars and 6 forks.
Related skills
Agent Observability
45j4flmao/agent-skills
Use when the user asks about agent observability, tracing agent reasoning, debugging an agent loop, OpenTelemetry GenAI spans for LLM calls, LLM cost tracking per agent, agent audit logs or hallucination/eval monitoring in production. Produces instrumentation config, trace schema, dashboards, alerts and cost attribution. Not for generic service tracing without LLMs (use devops-opentelemetry).
AI Agent Architectures
45j4flmao/agent-skills
Use when the user asks to build an AI agent, design an agent that calls tools and keeps memory, write an agent system prompt or pick an agent framework (Claude Agent SDK, OpenAI Agents SDK, LangGraph). Produces an architecture brief covering loop, tools, memory, guardrails and evals, routing depth to child skills. Not for workflow patterns (use ai-agentic-workflows).
AI Agent Core Loop
45j4flmao/agent-skills
Use when the user asks about the agent loop, ReAct vs plan-and-solve, re-planning, Reflexion, an agent stuck in a loop, max iterations or termination conditions. Produces a loop spec or patch with state, success predicate, caps and no-progress detection. Not for overall agent design (use ai-agent-architectures) or tool schemas (use tool-calling-principles).
Data Batch Processing
45j4flmao/agent-skills
Use when the user asks about Hive, HiveQL, Spark SQL batch jobs, Hive metastore, partitioning and bucketing, ORC vs Parquet for batch, Catalyst/AQE tuning or dynamic partition pruning. Produces batch job designs with table layout, file format and query tuning steps. Not for choosing Spark vs Dask vs Ray (use data-distributed-compute) or real-time pipelines (use data-streaming).
Cloud Native GCP Bigquery
45j4flmao/agent-skills
Use when the user asks about BigQuery cost, bytes scanned, partitioning and clustering, slots and editions, slow BigQuery queries, or Dremel and Capacitor internals. Produces table designs, query fixes and a pricing-model choice. Not for warehouse vendor comparison (use data-data-warehouse) or dbt projects (use data-science-analytics-engineering).
Data Clean Room
45j4flmao/agent-skills
Use when the user asks about a data clean room, AWS Clean Rooms, Snowflake Data Clean Rooms, private set intersection (PSI), privacy-preserving joins, differential privacy budgets or secure multi-party computation for data collaboration. Produces a clean-room design with analysis rules, PET choice and privacy budget. Not for in-house masking (use backend-data-masking).
Browse the most popular Claude skills, all databases skills, or write your own with the skills generator.