CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

graphql-schema-regression

Use when a pull request changes a GraphQL schema, or a Federation supergraph may have drifted. Diffs the schemas with graphql-inspector, classifies changes as breaking, dangerous or safe, applies rules such as deprecated-field removal and usage checks, and gates CI.

Quality

87%

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

The canonical home for this skill is testland/graphql-schema-regression

SKILL.md
Quality
Evals
Security

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.

A solid, actionable single-file skill with concrete commands, a rules reference, and ready-to-use CI configuration. Its main gaps are the unshown base-schema extraction in the Federation loop, a vague grep suggestion for machine parsing, and no remediation guidance when the CI gate fails.

Suggestions

Show how the per-subgraph base schemas are produced (e.g. a git show loop writing /tmp/${subgraph#subgraphs/}-base), since the Federation diff loop references files that are never created.

Replace "parse the human-readable output via grep on the classification headings" with the actual grep command and expected match for CI consumption.

Add a short "when the gate fails" step covering how to read the grouped breaking/dangerous output and remediate (revert, deprecate first, or add a --rule justification).

DimensionReasoningScore

Conciseness

Sections are command-first and lean, with genuinely non-obvious rationale like the Federation composite explanation. Minor over-explanation remains: the Overview prose restating what the tool is, the Output paragraph, and repeated "(Per [gqi-diff][diff])" citations. Efficient with minor trims possible fits 4 rather than the every-token-earns-its-place 5.

4 / 5

Actionability

Install, diff, introspect-then-diff, the rules flag table, the custom-rule export signature, and a full GitHub Actions YAML are copy-paste ready. Two gaps keep it from 5: the Federation loop diffs "/tmp/${subgraph#subgraphs/}-base" files whose creation is never shown, and "parse the human-readable output via grep" gives no concrete pattern.

4 / 5

Workflow Clarity

Sequences are clear (local diff, prod compare, CI extract-base then diff) and the validation checkpoint is explicit ("exits non-zero when breaking changes exist... the gate is the exit code"). Read-only diffs avoid the destructive/batch cap, but no error-recovery guidance exists for when the gate fails, so it sits at 4 rather than 5.

4 / 5

Progressive Disclosure

Well-organized sections with clearly signaled, one-level link definitions and a References section. At roughly 190 lines with no bundle files, the full CI YAML and Federation details are inlined where a references file could carry them, so it is good-but-not-fully-split structure (4) rather than the appropriately-split 5.

4 / 5

Total

16

/

20

Passed

Description

92%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.

A strong description: explicit trigger conditions, a named tool, and a complete concrete capability list in third-person voice. The only improvement space is broader natural-term coverage (e.g. "schema diff", "breaking changes", "SDL").

DimensionReasoningScore

Specificity

"Diffs the schemas with graphql-inspector, classifies changes as breaking, dangerous or safe, applies rules such as deprecated-field removal and usage checks, and gates CI" lists four concrete, tool-named actions with no coverage gaps. This matches the comprehensive-action anchor, not the minor-gaps anchor below it.

5 / 5

Completeness

An explicit "Use when..." clause answers when, and the diff/classify/rules/CI-gate sentence answers what, both concretely. This is the clear both-answered case, not the 4 anchor where the when could be more explicit.

5 / 5

Trigger Term Quality

Natural triggers like "a pull request changes a GraphQL schema" and "a Federation supergraph may have drifted" are present, but common variations such as "schema diff", "breaking changes", or "SDL" are missing. Good coverage with a few natural terms absent fits the 4 anchor rather than the synonym-complete 5.

4 / 5

Distinctiveness Conflict Risk

"GraphQL schema" and "Federation supergraph" carve a clear niche with distinct triggers and minimal conflict risk, distinguishing it from REST contract-diff skills. Third-person voice ("Diffs", "classifies", "applies", "gates") avoids the voice penalty.

5 / 5

Total

19

/

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

testland/qa

Path

plugins/qa-contract-testing/skills/graphql-schema-regression/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.