CtrlK
BlogDocsLog inGet started
Tessl Logo

airflow-new-sdk

Guide for implementing a brand-new language SDK for Airflow (AIP-108). Use this skill when a contributor wants to add support for a new programming language — designing the Python coordinator, implementing the wire protocol in the target language, writing the bundle format, and structuring the PR. Trigger on phrases like "new language SDK", "new SDK", "add support for [language]", "implement coordinator for", "SubprocessCoordinator", "BaseCoordinator", "new runtime", "AFBNDL01", "supervisor schema", or anything about bringing a new language into the Airflow executor ecosystem.

76

Quality

96%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

90%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.

A strong, lean overview skill: it routes contributors to the authoritative guide first, gives a genuinely executable base-class decision tree, and supplies exact paths, commands, and a PR checklist without any filler. The gaps are structural — the E2E section is mis-filed under Logging, and the workflow lacks explicit validation checkpoints between implementation steps.

Suggestions

Promote the 'E2E test suite' section out from under '## Logging' to its own top-level heading — it is a separate topic and its current placement makes navigation harder.

Add one or two explicit validation checkpoints to the workflow (e.g., after choosing a base class, confirm the coordinator skeleton passes the existing coordinator tests before implementing the wire protocol).

Consider moving the detailed log-parsing rules and E2E assertion-by-assertion design into a reference file (or the authoritative rst) to keep SKILL.md purely as the routing layer.

DimensionReasoningScore

Conciseness

The body assumes Claude's competence throughout — it never explains what Airflow, a coordinator, or a wire protocol is — and delivers everything via a decision tree, a reference table, file-layout blocks, and terse bullets. It sits at the "lean and efficient; every token earns its place" anchor; the only near-redundancy is repeated "read the rst first" reminders, which are arguably navigation signals rather than padding.

5 / 5

Actionability

Concrete, executable guidance throughout: exact repository paths for coordinator placement, a fully specified base-class decision tree, a copy-pasteable E2E test command, and a numbered PR checklist with exact file names. As an instruction-only skill its guidance is actionable without code, satisfying the rubric's code-vs-instruction note; the `<language>` placeholders are inherent to the task, not gaps.

5 / 5

Workflow Clarity

There is a clear implicit sequence (read the authoritative guide → pick a base class via the decision tree → study the matching reference implementation → implement logging → run E2E tests → complete the PR checklist), and the PR checklist is explicit. However, there are no validation checkpoints or error-recovery loops between steps (e.g., verify the bundle footer is accepted before moving on), keeping it at 'clear sequence with most checkpoints; minor validation gaps' rather than the score-5 anchor's feedback loops.

4 / 5

Progressive Disclosure

The skill is well structured as an overview that defers detail to the authoritative `30_new_language_sdk.rst` and to reference implementations, with references one level deep and clearly signaled. Two things hold it below the score-5 anchor: the '### E2E test suite' section is nested under '## Logging' despite being a separate topic, and moderately detailed material (log-record parsing rules, E2E assertion design) is inlined rather than split into a reference file.

4 / 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.

An exemplary description: it states what the skill does in concrete terms, gives an explicit 'Use when' clause, and enumerates natural trigger phrases including specific class and format identifiers. It is neither padded nor over-claiming.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions — "designing the Python coordinator, implementing the wire protocol in the target language, writing the bundle format, and structuring the PR" — which comprehensively cover the skill's scope. It clearly exceeds the score-4 anchor ("several specific actions; minor gaps") and matches the comprehensive multi-action anchor.

5 / 5

Completeness

Both 'what' ("Guide for implementing a brand-new language SDK for Airflow (AIP-108)" with the enumerated sub-tasks) and 'when' ("Use this skill when a contributor wants to add support for a new programming language" plus an explicit "Trigger on phrases like..." list) are clearly and explicitly stated, matching the score-5 anchor exactly.

5 / 5

Trigger Term Quality

Trigger phrases are comprehensive and include natural synonyms and jargon users would actually say: "new language SDK", "new SDK", "add support for [language]", "new runtime", plus class and format names ("SubprocessCoordinator", "AFBNDL01", "supervisor schema"). Nothing is generic; coverage matches the score-5 anchor including variations.

5 / 5

Distinctiveness Conflict Risk

The description occupies a clear niche (new language SDKs for the Airflow executor ecosystem) with distinct triggers like "AFBNDL01" and "SubprocessCoordinator" that no general skill would fire on. The trailing "anything about bringing a new language into the Airflow executor ecosystem" is broad but properly scoped to Airflow, keeping conflict risk minimal.

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.

Repository
apache/airflow
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.