write-script-duckdb

MUST use when writing DuckDB queries.

Quality

57%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./system_prompts/auto-generated/skills/write-script-duckdb/SKILL.md

Quality

Content

75%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 highly actionable with concrete commands and copy-paste SQL examples organized around clear intent-based workflows. It is somewhat verbose in restating the deploy distinction and inlines reference material rather than splitting it into bundle files.

Suggestions

Consolidate the repeated 'this is not a deploy' clarifications into one canonical statement to reduce redundancy and tighten the metadata-sync paragraph.

Add an explicit validation/side-effect checkpoint before running `wmill script preview` (e.g., confirm intended side effects, prefer --dry-run for generate-metadata) to strengthen the workflow feedback loop.

Consider extracting the DuckDB reference material (Ducklake, external DB, S3) into a one-level-deep reference file linked from SKILL.md to improve progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and actionable, but the deploy-vs-not distinction ("not a deploy") is restated across multiple sections and the metadata-sync paragraph repeats itself, fitting 'mostly efficient but includes some unnecessary explanation or could be tightened' rather than the lean 4-anchor.	3 / 5
Actionability	Guidance is fully executable: concrete wmill commands (preview, run, generate-metadata, rehash, --dry-run) and copy-paste-ready SQL covering arguments, Ducklake ATTACH, external DB resources, S3 read of csv/parquet/json, S3Object parameters, and COPY TO S3, matching the 'fully executable; covers common cases' anchor.	5 / 5
Workflow Clarity	A clear intent-based decision tree (preview vs run vs sync push) plus a metadata-sync workflow with a dry-run checkpoint and a diff-and-report feedback loop; not a 5 because execution-side validation for side-effecting previews is only implicitly the preview step itself.	4 / 5
Progressive Disclosure	Well-organized sections with clearly signaled external references (AGENTS.wmill.md, the preview skill, AGENTS.md) and no nested reference chains; not a 5 because the skill exceeds 50 lines and inlines substantial DuckDB reference material that, while reasonable, is not split into one-level-deep bundle files.	4 / 5
	Total	16 / 20 Passed

Description

40%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 functions as a clear trigger condition but omits any statement of what the skill does, leaving it as a 'when-only' description. Its DuckDB niche is distinct, but capability specificity and completeness are weak.

Suggestions

Add an explicit 'what' clause listing concrete capabilities, e.g. 'Writes DuckDB scripts with $name argument syntax, Ducklake attaches, external database connections, and S3 read/write operations.'

Broaden trigger terms to include natural synonyms users might say, such as 'SQL', 'ducklake', or 'query S3/Parquet files in DuckDB'.

Combine what and when into a single sentence mirroring the rubric's good example: concrete actions followed by 'Use when writing or running DuckDB queries...'.

Dimension	Reasoning	Score
Specificity	The description names the domain ("writing DuckDB queries") but offers only a single minimal action and no enumeration of what the skill actually does, matching the 'names the domain but actions are minimal or generic' anchor rather than the 3-anchor which requires 1-2 concrete capability actions.	2 / 5
Completeness	The text is purely a 'when' clause ("MUST use when writing DuckDB queries") with no statement of what the skill does, which is the 'only when is present without what' case; it does not reach 3 because there is no clear 'what'.	2 / 5
Trigger Term Quality	"writing DuckDB queries" supplies the natural keyword a user would say, but coverage is limited to DuckDB/queries and omits common synonyms or related terms (SQL, ducklake, S3, external databases), fitting the 'some relevant keywords but missing common variations' anchor.	3 / 5
Distinctiveness Conflict Risk	"DuckDB queries" carves out a fairly distinct niche with minimal conflict risk, though it could overlap with sibling SQL-dialect skills, placing it at 'mostly distinct; minor overlap risk' rather than the fully distinct 5-anchor.	4 / 5
	Total	11 / 20 Passed

Validation

100%

Warnings & errors only

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: windmill-labs/windmill
Path: system_prompts/auto-generated/skills/write-script-duckdb/SKILL.md
Commit: 74af4ed

Reviewed: about 18 hours ago

Table of Contents

Discovery Implementation Validation

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.