CtrlK
BlogDocsLog inGet started
Tessl Logo

cosmos-dbt-fusion

Run a dbt Fusion project with Astronomer Cosmos. Use when running a dbt Fusion project with Astronomer Cosmos (Cosmos 1.11+, ExecutionMode.LOCAL on Snowflake/Databricks). Before implementing, verify dbt engine is Fusion (not Core), the warehouse is supported, and local execution is acceptable. Does not cover dbt Core.

66

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

78%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, executable implementation checklist with a clear numbered workflow, per-step validation in the setup steps, and a final validation plus user-test section. The main defects are a dangling reference to a nonexistent reference/cosmos-config.md file (twice), some repeated version/config content, and a thin troubleshooting section.

Suggestions

Add the referenced reference/cosmos-config.md file (or remove both links to it) so the advertised ProfileConfig/operator_args details actually exist in the bundle.

Deduplicate the Cosmos 1.11.0 version note (frontmatter note, intro, and step 1 all restate it) and have Option A reference the configs from steps 4–6 instead of re-inlining them.

Expand the Troubleshooting section: explain what AIRFLOW__COSMOS__PRE_DBT_FUSION=1 does and add a diagnose-fix-retry loop instead of a bare env-var snippet.

DimensionReasoningScore

Conciseness

The body is directive rather than explanatory — tables of constraints, code, and validation steps with no padding about what dbt/Cosmos/Airflow are. Minor over-explanation and repetition remain: the Cosmos 1.11.0 version note appears three times (frontmatter note, intro, step 1) and Option A re-inlines the profile/execution configs already shown in steps 4–5, which fits the efficient-with-minor-trimmables anchor rather than the every-token-earns-its-place anchor at 5.

4 / 5

Actionability

Guidance is fully executable: `pip install "astronomer-cosmos>=1.11.0"`, a concrete Dockerfile install with `curl -fsSL https://public.cdn.getdbt.com/fs/install/install.sh | sh -s -- --update`, common install paths, and complete copy-paste-ready DbtDag and DbtTaskGroup examples covering both common integration cases. The bare `AIRFLOW__COSMOS__PRE_DBT_FUSION=1` troubleshooting snippet is a very minor gap that does not undermine the otherwise executable coverage.

5 / 5

Workflow Clarity

Steps 1–8 are clearly numbered and sequenced, steps 1 and 2 carry explicit `**Validate**` checkpoints, and step 8 consolidates a final checklist plus "User Must Test" items. It falls short of the anchor-5 explicit feedback-loop pattern because steps 3–7 have no inline checkpoints and the troubleshooting section offers a single unexplained env var rather than a diagnose-fix-retry loop.

4 / 5

Progressive Disclosure

The body twice points to "reference/cosmos-config.md" with clear signaling, but no bundle directory exists and that referenced file is absent, so the content split it advertises is not real and navigation dead-ends. The inline body itself is well-sectioned, so this fits the some-structure-with-flawed-references anchor rather than the minimal-structure anchor at 2 or the good-structure anchor at 4.

3 / 5

Total

16

/

20

Passed

Description

82%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 explicitly answers both what and when with concrete, version-gated trigger conditions, and explicitly scopes itself against dbt Core to avoid conflict with the sibling skill. The only notable weakness is that the capability list stays at the level of "run + verify" without naming the concrete implementation actions (install binary, configure, assemble DAG) that make up the skill.

DimensionReasoningScore

Specificity

"Run a dbt Fusion project with Astronomer Cosmos" names the domain and one concrete capability, plus verify preconditions ("verify dbt engine is Fusion (not Core), the warehouse is supported, and local execution is acceptable"). Operational capabilities like installing the Fusion binary, configuring Cosmos, or assembling DAGs are absent, matching "1-2 concrete actions, but not comprehensive" rather than the several-specific-actions anchor at 4.

3 / 5

Completeness

Both parts are explicit: "Run a dbt Fusion project with Astronomer Cosmos" answers what, and "Use when running a dbt Fusion project with Astronomer Cosmos (Cosmos 1.11+, ExecutionMode.LOCAL on Snowflake/Databricks)" answers when with concrete trigger conditions. This matches the clearly-and-explicitly-both anchor; the 4 anchor would require the 'when' clause to be less specific, which it is not.

5 / 5

Trigger Term Quality

"dbt Fusion", "Astronomer Cosmos", "Snowflake", "Databricks", and "ExecutionMode.LOCAL" are exactly the phrases a user would naturally say when needing this skill. A few natural trigger terms are missing (e.g. "Airflow", "DAG", warehouse names BigQuery/Redshift), so it fits the good-coverage-with-few-missing anchor rather than the comprehensive-synonym anchor at 5.

4 / 5

Distinctiveness Conflict Risk

"Does not cover dbt Core" explicitly disambiguates from the sibling cosmos-dbt-core skill, and the Fusion-engine/warehouse/execution-mode conditions carve out a clear niche with minimal conflict risk. This matches the clear-niche-with-distinct-triggers anchor.

5 / 5

Total

17

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 2 missing

Warning

Total

15

/

16

Passed

Repository
astronomer/agents
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.