CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

write-script-graphql

MUST use when writing GraphQL queries.

Quality

64%

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

Fix and improve this skill with Tessl

tessl review fix ./system_prompts/auto-generated/skills/write-script-graphql/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

71%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 exact commands, copy-paste-ready GraphQL examples, and a well-gated deploy/test/metadata workflow with real validation checkpoints. Its weaknesses are redundancy (the deploy-gating rule restated several times) and a monolithic single-file structure that inlines a general Windmill script-writing guide not specific to GraphQL. Trimming the repetition and splitting the general guide into a shared reference would lift both.

Suggestions

State the deploy-only-on-explicit-request rule once (in the CLI commands list) and reference it from other sections instead of restating it four times; likewise collapse the three "this is not a deploy" clarifications into one.

Move the general Windmill script-writing guide (resource-typed parameters, return values, preprocessor scripts — none of it GraphQL-specific) into a shared reference file, leaving SKILL.md focused on the GraphQL script workflow.

Reorder so the GraphQL-writing guidance and the post-writing workflow (preview → metadata → deploy) read as one continuous sequence instead of interleaving the CLI section, general principles, and GraphQL guide.

DimensionReasoningScore

Conciseness

The content is Windmill-specific with no generic padding, but has real redundancy: the deploy-only-on-explicit-request rule is stated four times (CLI list, "Preview vs run", "Only use sync push when", and the preview section), "not a deploy" is repeated three times, and the AGENTS.wmill.md "Deploying" pointer appears twice. Mostly efficient but could be tightened by consolidating these — matching "includes some unnecessary explanation or could be tightened".

3 / 5

Actionability

Fully executable throughout: exact CLI commands (`wmill script preview <script_path>`, `wmill generate-metadata --dry-run`, `wmill generate-metadata rehash`, `wmill resource-type list --schema`), copy-paste-ready GraphQL examples covering queries, variable binding, and mutations, and concrete per-language placeholder syntax. Specific examples cover the common cases, matching the top anchor.

5 / 5

Workflow Clarity

A clear script lifecycle is sequenced (write → choose preview vs run by intent → generate-metadata → deploy only on explicit request) with explicit validation checkpoints: preview validates before any deploy, regenerated `.lock`/`.script.lock` diffs are surfaced to catch version bumps, and `--dry-run` lists stale items before narrowing. The destructive operation (deploy) does have validation, so no cap applies; it is 4 rather than 5 because the end-to-end workflow is interleaved across separated sections and must be assembled by the reader.

4 / 5

Progressive Disclosure

The single 123-line file has well-organized headers and no nested references (the AGENTS.wmill.md pointer is one level and clearly signaled), but roughly half the content — the general Windmill script-writing guide on resource types, return values, and preprocessor scripts — is not GraphQL-specific and would naturally live in a separate shared reference file. This matches "Some structure but could be better organized; content that should be separate is inline"; the under-50-line simple-skill exception does not apply at this length.

3 / 5

Total

15

/

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 provides a clear, natural trigger but is missing the entire "what" half — it never states what the skill does (write queries and mutations with bound variables for Windmill scripts). It also omits key trigger terms like "mutations" and "GraphQL API". Expanding it to one or two sentences covering capabilities would move it into good-example territory.

Suggestions

Add a "what" clause listing concrete capabilities, e.g. "Write GraphQL queries and mutations with query-parameter arguments and automatically bound variables, as Windmill scripts. Use when writing GraphQL queries, mutations, or interacting with GraphQL APIs."

Include the missing natural trigger term "mutations" (the body covers them but the description never mentions it), plus variations like "GraphQL API" or "GraphQL endpoint".

State the concrete actions (queries, mutations, variable binding) so the description also distinguishes this skill from generic SQL/database query skills that share the word "queries".

DimensionReasoningScore

Specificity

"MUST use when writing GraphQL queries" names the domain but lists no concrete actions or capabilities, matching the anchor example "Processes PDF files". It is not 3 because no specific capabilities (queries, mutations, variable binding) are mentioned, and not 1 because the domain is explicitly named rather than pure abstraction.

2 / 5

Completeness

Only the "when" is present ("MUST use when writing GraphQL queries") with no "what" — the description never states what the skill actually does. This exactly matches the anchor "only 'when' is present without 'what'" (example: "Use when working with documents").

2 / 5

Trigger Term Quality

"GraphQL" and "queries" are natural terms users would say, but common variations are missing — notably "mutations" (a use case the body itself covers), "GraphQL API", and "schema". This matches "Some relevant keywords but missing common variations or synonyms".

3 / 5

Distinctiveness Conflict Risk

"GraphQL" is a clear niche trigger unlikely to fire for the wrong skill, but the generic term "queries" carries minor overlap risk with SQL/database query-writing skills. This fits "Mostly distinct; minor overlap risk with closely related skills" rather than 5.

4 / 5

Total

11

/

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

windmill-labs/windmill

Path

system_prompts/auto-generated/skills/write-script-graphql/SKILL.md

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.