write-script-bash

MUST use when writing Bash scripts.

Quality

51%

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-bash/SKILL.md

Quality

Content

67%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 delivers concrete, Windmill-specific operational guidance with executable Bash examples and a clear preview/run/metadata workflow supported by validation checkpoints. Its main weakness is repetition of the deploy-safety caveat and a dense inline metadata section that could be extracted to a reference.

Suggestions

Consolidate the repeated 'not a deploy / only deploy when explicitly asked' caveat into a single prominent callout instead of restating it across the CLI list, preview-vs-run, metadata, and after-writing sections.

Make the preview-failure recovery loop explicit (e.g., 'If preview errors, fix the script and re-preview before any deploy') to lift workflow_clarity to a 5.

Consider moving the detailed generate-metadata rehash/dry-run/strict-folder guidance into a references file linked from a short summary, improving both conciseness and progressive_disclosure.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient Windmill-specific operational detail Claude would not know, but it restates the deploy safety distinction ('not a deploy', 'only when the user explicitly asks to deploy/publish/push') three to four times, which is noticeable padding that could be tightened.	3 / 5
Actionability	Concrete wmill commands with flags (preview, run, generate-metadata, generate-metadata rehash, --dry-run) and executable Bash examples for structure, JSON output, and env vars provide mostly copy-paste-ready guidance with only minor arg-format gaps.	4 / 5
Workflow Clarity	The preview-vs-run decision tree is clearly sequenced by intent, and metadata sync includes checkpoints (dry-run to list stale items, diff the regenerated .lock to surface version bumps, gated deploy); it falls short of 5 because the recovery loop on a failed preview is implicit rather than spelled out.	4 / 5
Progressive Disclosure	Content is well-organized under clear headers with one-level-deep inline references to AGENTS.wmill.md and the preview skill and no nested references; it does not reach 5 because the dense metadata-sync guidance is inlined rather than split into a reference file despite exceeding the simple-skill line budget.	4 / 5
	Total	15 / 20 Passed

Description

36%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 a bare trigger clause with no capability statement: it tells Claude when to fire but not what the skill does or that it is Windmill-specific. This risks triggering on ordinary bash work and leaves the what/when contract half-empty.

Suggestions

Add a 'what' clause stating the concrete actions, e.g. 'Write and test Windmill Bash scripts: structure positional-argument inputs, emit JSON output, and sync metadata with wmill generate-metadata.'

Qualify the trigger with the Windmill context to avoid conflicts, e.g. 'Use when writing Bash scripts for a Windmill workspace' and include synonyms like 'shell scripts' or '.sh'.

Mirror the structure of the good examples: concrete capabilities first, then an explicit 'Use when...' clause naming natural trigger phrases.

Dimension	Reasoning	Score
Specificity	The phrase 'writing Bash scripts' names the domain but describes no concrete actions the skill performs, matching the 'names the domain but actions are minimal or generic' anchor rather than the entirely-vague anchor 1.	2 / 5
Completeness	Only the 'when' is present ('when writing Bash scripts') with no statement of what the skill does, matching the 'only when is present without what' anchor; it does not reach 3 because the what is entirely missing rather than weakly implied.	2 / 5
Trigger Term Quality	'writing Bash scripts' is a natural phrase a user would say, but synonyms and variations (shell scripts, shell scripting, .sh files) are absent, matching the 'some relevant keywords but missing common variations' anchor.	3 / 5
Distinctiveness Conflict Risk	'MUST use when writing Bash scripts' lacks any Windmill qualifier, so it would trigger for any bash scripting and overlaps with generic shell-scripting skills; it is more specific than anchor 2 but does not reach the low-conflict 4 anchor.	3 / 5
	Total	10 / 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-bash/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.