GraphQL Schema Analyzer skill for Claude and Codex
An audit of an existing GraphQL schema: an inventory of types and operations, the gaps a client team will hit (missing descriptions, nullable-everything…
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 an existing GraphQL schema needs an audit for client-facing gaps like missing descriptions, unpaginated lists, loose nullability, and stale deprecations before more clients build on it. Do not use for designing a schema from scratch, fixing a schema that fails to load, or writing end-user docs. Related topics: coding, graphql.
What it will ask you for
- Schema SDL (longer free text)
- Focus areas (choose any: Types, Queries, Mutations, Subscriptions, Directives, Scalars, Pagination, Nullability)
- Depth of analysis (choose one: Overview, Detailed, Exhaustive)
- Highlight deprecated fields (choose one: Yes, No)
- Who consumes it (choose one: One web client, Web and mobile clients, Public API with external developers, Internal services)
- Output format (choose one: Markdown report, JSON summary, HTML page)
How it works
- Inventory the schema: counts per kind, the root operations, and the entry points clients will use most.
- Assess each selected focus area and record every finding with the type and field it concerns.
- List each @deprecated field with its reason and replacement status when deprecations are on.
- Weight the findings by the consumer type so a public API's breaking risks rank above an internal service's cosmetic ones.
- Write the recommendations ranked by impact and cost, marking which are breaking changes, sized to the chosen depth.
What you get
A report in the chosen format with sections Summary, Inventory, Findings (grouped by focus area with type and field references), Deprecations when requested, and Recommendations ranked with a breaking flag.
What it watches out for
- Flagging every nullable field as a problem when GraphQL nullability is meant to express partial failure, not optional data.
- Reporting counts and structure without saying what a client would do differently, which reads as a listing rather than an audit.
- Recommending breaking renames to a public schema without an accompanying deprecation path.
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.