airflow-dag-patterns
Build production Apache Airflow DAGs with best practices for operators, sensors, testing, and deployment. Use when creating data pipelines, orchestrating workflows, or scheduling batch jobs.
78
71%
Does it follow best practices?
Impact
90%
0.95xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./tests/ext_conformance/artifacts/agents-wshobson/data-engineering/skills/airflow-dag-patterns/SKILL.mdQuality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong skill description that concisely covers specific capabilities (DAGs, operators, sensors, testing, deployment), includes a clear 'Use when' clause with natural trigger terms, and is distinctly scoped to Apache Airflow. It uses proper third-person voice and avoids vague language or unnecessary verbosity.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: building DAGs, using operators, sensors, testing, and deployment. Also mentions best practices, which adds practical context. | 3 / 3 |
Completeness | Clearly answers both 'what' (build production Apache Airflow DAGs with best practices for operators, sensors, testing, and deployment) and 'when' (explicit 'Use when' clause covering data pipelines, orchestrating workflows, or scheduling batch jobs). | 3 / 3 |
Trigger Term Quality | Includes strong natural keywords users would say: 'Airflow', 'DAGs', 'data pipelines', 'orchestrating workflows', 'scheduling batch jobs', 'operators', 'sensors'. These cover common variations of how users would describe this need. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive due to the specific mention of 'Apache Airflow' and 'DAGs', which clearly carve out a niche. Unlikely to conflict with general workflow or pipeline skills due to the technology-specific framing. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
42%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides highly actionable, production-quality Airflow code examples covering a comprehensive range of patterns. However, it is severely over-long and monolithic—cramming ~400 lines of detailed patterns into a single file without progressive disclosure. The content would benefit greatly from splitting patterns into separate files and trimming the SKILL.md to a lean overview with references.
Suggestions
Split each pattern (TaskFlow, Dynamic DAGs, Branching, Sensors, Error Handling, Testing) into separate referenced files (e.g., patterns/taskflow.py, patterns/sensors.py) and keep only the Quick Start example and a pattern index in SKILL.md.
Remove the DAG Design Principles table and Do's/Don'ts section—these are well-known Airflow best practices that Claude already knows; at most, reduce to a 3-line reminder of the most non-obvious points.
Add an explicit development workflow with validation checkpoints: write DAG → run `python dags/my_dag.py` to check parse errors → run pytest → deploy → verify in Airflow UI.
Trim repeated boilerplate across examples (e.g., default_args, identical imports) by defining them once and referencing them in subsequent patterns.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~400+ lines. The DAG design principles table explains concepts Claude already knows (idempotency, atomicity). Multiple full DAG examples repeat boilerplate (imports, default_args, DAG instantiation). The do's/don'ts section restates common knowledge. Much of this could be condensed to key patterns with minimal scaffolding. | 1 / 3 |
Actionability | All code examples are complete, executable Python files with proper imports, realistic configurations, and copy-paste ready patterns. The testing section includes runnable pytest examples, and the sensor/error handling patterns include concrete configurations with specific timeout values and modes. | 3 / 3 |
Workflow Clarity | Individual patterns are well-structured with clear task dependency chains, and the testing pattern provides validation steps. However, there's no overarching workflow for creating/deploying a DAG (e.g., write → test → deploy → validate in production), and no explicit validation checkpoints or feedback loops for the development process itself. | 2 / 3 |
Progressive Disclosure | Monolithic wall of content with six full pattern implementations inline. No bundle files exist to offload detailed patterns. The project structure section, testing examples, and individual patterns (dynamic DAGs, branching, sensors, error handling) should be in separate referenced files, with SKILL.md serving as a concise overview pointing to them. | 1 / 3 |
Total | 7 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (526 lines); consider splitting into references/ and linking | Warning |
Total | 10 / 11 Passed | |
- Repository
- Dicklesworthstone/pi_agent_rust
- Commit
99da384
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.