CtrlK
BlogDocsLog inGet started
Tessl Logo

wagneripjr/airflow-dags

Author, debug and configure Apache Airflow 3 DAGs: TaskFlow API, scheduling and assets, XCom, sensors, dynamic task mapping, and multi-layer test suites

94

1.46x
Quality

94%

Does it follow best practices?

Impact

97%

1.46x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No findings from the security scan

Overview
Quality
Evals
Security
Files

Quality

Content

86%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is well-structured and highly actionable, with executable code and a strong reference architecture that keeps the overview lean while delegating depth to real bundle files. Its only weakness is mild verbosity in the template and a lack of explicit feedback-loop checkpoints in the main flow.

Suggestions

Tighten the TaskFlow DAG Template: trim obvious inline comments (e.g. '# API hammered', '# I/O on every parse') and consider shortening the template to its essential, copy-paste-ready core to lift conciseness.

Add an explicit validate→fix→retry feedback loop to the authoring workflow (e.g. after drafting, run dag.test() then structural validation, fix, re-test) so the main flow surfaces its checkpoints rather than leaving them implicit in the template.

Condense the restated principles and a few redundant Common Mistakes rows that overlap the Critical Rules to further reduce token cost.

DimensionReasoningScore

Conciseness

Mostly efficient — decision table, terse BAD/GOOD pairs, and a reference list assume Claude's intelligence — but the full template and a few restatable comments ('API hammered') could be trimmed, keeping it just below lean.

4 / 5

Actionability

Provides a complete, copy-paste-ready TaskFlow template with concrete default_args (retries, backoff, timeout), specific settings (AIRFLOW__OPERATORS__DEFAULT_DEFERRABLE=True), and BAD/GOOD code covering the common cases.

5 / 5

Workflow Clarity

Clear sequence (Quick Decision Guide → Critical Rules → Template → Common Mistakes → References) with a validation checkpoint via the dag.test() block, though explicit validate→fix→retry feedback loops are not surfaced in the main authoring flow.

4 / 5

Progressive Disclosure

Clear overview SKILL.md with a Quick Decision Guide and Reference Files section mapping needs to 12 one-level-deep, real bundle files, all clearly signaled via markdown links.

5 / 5

Total

18

/

20

Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is third-person, concise yet comprehensive, and clearly answers both what the skill does and when to use it. It lists many concrete actions and a strong set of natural trigger terms with minimal conflict risk.

DimensionReasoningScore

Specificity

Lists multiple specific concrete actions — 'builds data pipelines with TaskFlow API or traditional operators, configures scheduling and asset-driven triggers, wires XCom data passing, sets up sensors and deferrable operators, generates dynamic task mappings, and structures multi-layer test suites' — with comprehensive coverage.

5 / 5

Completeness

Explicitly answers both 'what' (the action list) and 'when' ('Use when creating, debugging, or configuring Apache Airflow 3 DAGs' plus the 'Triggers on…' clause), matching the top anchor.

5 / 5

Trigger Term Quality

Comprehensive natural trigger terms a user would actually say — 'DAG authoring, TaskFlow API, operators, sensors, scheduling, assets, dynamic tasks, XCom, or pipeline testing' — covering the Airflow vocabulary; no file extensions apply to this topic.

5 / 5

Distinctiveness Conflict Risk

Clear Airflow-3-DAG niche with distinct, domain-specific triggers unlikely to collide with other skills; the airflow-dags name reinforces specificity.

5 / 5

Total

20

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Reviewed

Table of Contents