Skip to main content
skill CX Codex

graphql-and-hidden-parameters — Claude Code skill: GraphQL testing

A playbook for GraphQL and hidden parameter testing: introspection, batching, undocumented fields and authorization gaps.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add yaklang/hack-skills --skill graphql-and-hidden-parameters -a claude-code
Open on GitHub

What graphql-and-hidden-parameters does

The skill sets an order of work for projects that expose GraphQL or whose REST docs hint at optional and deprecated fields. First pass is introspection: __typename and __schema queries that list types. When introspection is locked down, it moves to field suggestions from error messages, known type probes such as __type(name: "User"), and route extraction from JS and mobile bundles.

Then comes a table of high-value checks: IDOR through user(id: "victim"), batching an array of login or object-fetch operations, admin-only fields that leak into type definitions, and nested objects guarded by weaker checks. Batching is the force multiplier here — one operation pulls a lot of checks with it.

Hidden parameter discovery leans on mismatches: fields in admin docs but not public ones, permissive schemas with additionalProperties, frontend code sending richer request bodies than the UI controls suggest, mobile endpoints carrying role, org, feature-flag or internal filter fields. The skill routes onward to companion skills for API authorization, JWT abuse and endpoint recon. Use it on your own applications and authorized engagements.

What it helps with

  • check whether GraphQL introspection is left open
  • find admin-only fields exposed in the schema
  • collect hidden parameters from mobile endpoints
  • test authorization on nested object fields
  • see whether batching changes rate or auth behavior

Example prompts for the agent

  • Check GraphQL introspection on our staging host and list the types
  • Find fields in the schema that a regular user should not reach
  • Compare frontend and mobile request bodies — which parameters appear only in the mobile one?

How to install graphql-and-hidden-parameters

1Skills CLI (Vercel)

npx skills add yaklang/hack-skills --skill graphql-and-hidden-parameters -a claude-code

2Manually: put the files in place

~/.claude/skills/graphql-and-hidden-parameters/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/graphql-and-hidden-parameters/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about graphql-and-hidden-parameters

How do I install graphql-and-hidden-parameters in Claude?

Run npx skills add yaklang/hack-skills --skill graphql-and-hidden-parameters -a claude-code. The skill lands in ~/.claude/skills/graphql-and-hidden-parameters/SKILL.md.

What does graphql-and-hidden-parameters do?

It walks through GraphQL testing: introspection, batching, undocumented fields, hidden parameters and authorization gaps. Each step comes with concrete queries and pointers on where to look next.

Which skills does it hand off to?

When hidden fields affect privilege it points to api-authorization-and-bola. If batching shifts auth or rate behavior, it routes to api-auth-and-jwt-abuse; for incomplete endpoint discovery, to api-recon-and-docs.

Source text

SKILL.md · MIT
---
name: graphql-and-hidden-parameters
description: >-
  GraphQL and hidden parameter testing playbook. Use when exploring introspection, batching, undocumented fields, hidden parameters, schema abuse, and GraphQL authorization gaps.
---

# SKILL: GraphQL and Hidden Parameters — Introspection, Batching, and Undocumented Fields

> **AI LOAD INSTRUCTION**: Use this skill when GraphQL exists or when REST documentation suggests optional, deprecated, or undocumented fields. Focus on schema discovery, hidden parameter abuse, and batching as a force multiplier.

## 1. GRAPHQL FIRST PASS

```graphql
query { __typename }
query {
  __schema {
    types { name }
  }
}
```

If introspection is restricted, continue with:

- field suggestions and error-based discovery
- known type probes like `__type(name: "User")`
- JS and mobile bundle route extraction

## 2. HIGH-VALUE GRAPHQL TESTS

| Theme | Example |
|---|---|
| IDOR | `user(id: "victim")` |
| batching | array of login or object fetch operations |
| hidden fields | admin-only fields exposed in type definitions |
| nested authz gaps | related object fields with weaker checks |

## 3. HIDDEN PARAMETER DISCOVERY

Look for:

- fields present in admin docs but not public docs
- `additionalProperties` or permissive schemas
- frontend code using richer request bodies than visible UI controls
- mobile endpoints carrying role, org, feature-flag, or internal filter fields

## 4. NEXT ROUTING

- If hidden fields affect privilege: [api authorization and bola](../api-authorization-and-bola/SKILL.md)
- If GraphQL batching changes auth or rate behavior: [api auth and jwt abuse](../api-auth-and-jwt-abuse/SKILL.md)
- If endpoint discovery is incomplete: [api recon and docs](../api-recon-and-docs/SKILL.md)

© yaklang · MIT · yaklang/hack-skills

Similar skills

All skills for Security →

From the same set93

kernel-exploitation A Linux kernel security research playbook for AI coding assistants. It covers QEMU debugging environments and mitigation mechanisms like KASLR and SMEP. CX Codex +1 3,600installs request-smuggling This skill tests HTTP desynchronization between front proxies and origin servers. Use it for CL.TE, TE.CL, HTTP/2-to-HTTP/1, and client-side desync checks. CX Codex +1 3,600installs file-access-vuln A routing skill for testing download endpoints, file paths, LFI and upload flows. It points the agent at the right topic skill. CX Codex +1 3,600installs xxe-xml-external-entity Guides your agent through XXE vulnerability auditing across XML, SVG, SOAP, and Office document parsing pipelines. CX Codex +1 3,600installs ssti-server-side-template-injection An SSTI playbook: find server-side template injection, fingerprint the engine, and follow the chain to command execution. CX Codex +1 3,600installs binary-protection-bypass A skill that walks through ELF binary protections and the bypass for each one. Useful for CTF work and exploit analysis. CX Codex +1 3,700installs cmdi-command-injection This skill helps find command injection where user input reaches shells, process execution, converters, import pipelines, or blind OOB sinks. CX Codex +1 3,700installs anti-debugging-techniques Anti-debugging detection and bypass playbook for protected Linux and Windows binaries. Helps reverse engineers handle ptrace, PEB flags, execution timing checks, and thread hiding. CX Codex +1 3,700installs

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor