GraphQL Schema Validator skill for Claude and Codex
A pre-merge validation report for a GraphQL schema: each rule in your chosen ruleset checked and every violation listed with line number and severity, plus…
Install it once and Claude or Codex runs it for you. When your request matches, your AI loads the skill, asks for anything it is missing, and works through the template itself. Prefer to fill it in by hand? See the full prompt template.
Download the skill (.zip) · View SKILL.md
When it is used
Use when a GraphQL schema and the queries clients send need checking against lint rules and the spec before a merge, with every violation listed by line and severity. Do not use for diagnosing one failing error from server logs, or a broad design audit with recommendations. Related topics: coding, graphql.
What it will ask you for
- Schema SDL (longer free text)
- Sample queries (longer free text)
- Validation ruleset (choose any: Type consistency, Field existence, Directive usage, Deprecation warnings, Naming conventions, Required descriptions, Input type cycles)
- Severity threshold (choose one: All, Warnings only, Errors only)
- Output format (choose one: Markdown, JSON, Plain text)
- Include suggested fixes (choose one: Yes, No)
How it works
- Check spec validity first: defined types, no duplicates, no input non-null cycles, directives on declared locations, interfaces fully implemented.
- Apply each selected rule and record a finding with line number, rule, and severity per violation.
- Validate every sample query against the schema for unknown fields, argument errors, bad selections, wrong fragment types, and deprecated field use.
- Filter by the severity threshold and sort errors first then by line, adding a one-line fix per finding when fixes are on.
- State the pass or fail verdict with counts per severity.
What you get
A report in the chosen format: a verdict line with counts, then a findings table (line, rule, severity, message, fix when enabled) for the schema followed by one per sample query.
What it watches out for
- Reporting style violations before spec errors, so a schema that cannot load still looks mostly green.
- Marking deprecated field use in a query as an error when it is a warning the client may knowingly accept.
- Skipping the sample queries because the schema passed, when the queries are where production breaks.
How to install
Claude.ai and the Claude desktop app
- Download the skill .zip.
- Open Settings, then Capabilities, and find Skills.
- Choose Upload skill and select the .zip.
- Ask Claude for help with the task. It loads the skill when your request matches.
Claude Code
- Download the skill .zip.
- Unzip it into ~/.claude/skills/ (or .claude/skills/ inside a project).
- Start a new session. Claude Code loads the skill when your request matches.
Codex
- Download the skill .zip.
- Unzip it into ~/.agents/skills/ (or .agents/skills/ inside a project).
- Codex picks it up on its own. Type $ and the skill name to use it, or just ask for the task.
Cursor
- Download the Cursor rule (.mdc).
- Save it in .cursor/rules/ inside your project.
- Cursor's agent applies the rule when your request matches its description.
AGENTS.md (Codex, Cursor, and other coding agents)
- Copy the AGENTS.md section.
- Paste it into AGENTS.md at the root of your repository, creating the file if needed.
- Coding agents that read AGENTS.md follow it when your request fits.