CtrlK
BlogDocsLog inGet started
Tessl Logo

airflow-operator-creator

Airflow Operator Creator - Auto-activating skill for Data Pipelines. Triggers on: airflow operator creator, airflow operator creator Part of the Data Pipelines skill category.

36

1.02x

Quality

3%

Does it follow best practices?

Impact

96%

1.02x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./planned-skills/generated/11-data-pipelines/airflow-operator-creator/SKILL.md
SKILL.md
Quality
Evals
Security

Evaluation results

99%

-1%

REST API Data Ingestion Operator

Production-ready custom Airflow ETL operator

Criteria
Without context
With context

BaseOperator subclass

100%

100%

execute() method

100%

100%

Parameterized operator

100%

100%

Error handling

100%

100%

Logging present

100%

87%

DAG default_args retries

100%

100%

DAG schedule configured

100%

100%

XCom or return value

100%

100%

template_fields defined

100%

100%

No hardcoded values in operator

100%

100%

retry_delay configured

100%

100%

README documents parameters

100%

100%

Without context: $0.5015 · 2m 2s · 30 turns · 30 in / 6,515 out tokens

With context: $0.4296 · 1m 26s · 26 turns · 59 in / 5,255 out tokens

93%

8%

Sales Data Transformation Pipeline

Data transformation DAG with output validation

Criteria
Without context
With context

Validation task present

100%

100%

Validation fails pipeline

100%

100%

Duplicate removal

100%

100%

Invalid record filtering

70%

70%

Aggregation by category

100%

100%

DAG retries configured

100%

100%

Task dependencies defined

100%

100%

Logging in tasks

0%

100%

Summary CSV produced

100%

100%

Default_args includes owner and start_date

50%

50%

No hardcoded absolute paths

100%

100%

Without context: $0.4507 · 1m 47s · 22 turns · 64 in / 7,369 out tokens

With context: $0.5671 · 2m 1s · 29 turns · 29 in / 8,077 out tokens

98%

Nightly Inventory Sync Pipeline

Workflow orchestration with step-by-step DAG design

Criteria
Without context
With context

Four or more tasks

100%

100%

Explicit task dependencies

100%

100%

Retry configuration

100%

100%

Failure isolation

100%

90%

Step-by-step design notes

100%

100%

Logging in tasks

100%

100%

Data merge/dedup step

100%

100%

Outcome notification task

100%

100%

DAG catchup disabled or configured

100%

100%

No hardcoded absolute paths

100%

100%

Default_args start_date and owner

60%

80%

DAG id meaningful

100%

100%

Without context: $0.8788 · 3m 52s · 41 turns · 39 in / 12,655 out tokens

With context: $0.8305 · 3m 25s · 38 turns · 176 in / 13,795 out tokens

Repository
jeremylongshore/claude-code-plugins-plus-skills
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

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.