OpenParable
Skip to content
← Back to skills

Wstg Apit 99

Testing & Quality
wstg-apit-99

Wstg Apit 99 is an Agent Skill for testing & quality published in the CyberStrikeus/CyberStrike repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Testing GraphQL

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/CyberStrikeus/CyberStrike/blob/HEAD/.cyberstrike/skill/WEB/OWASP_WSTG_4.2/wstg-apit-99/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/wstg-apit-99, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

wstg-apit-99

Test ID

WSTG-APIT-99

Test Name

Testing GraphQL APIs

High-Level Description

GraphQL is a query language for APIs that allows clients to request exactly the data they need. While flexible, GraphQL APIs introduce unique security challenges including introspection, batching attacks, nested queries, and authorization bypass opportunities.


What to Check

  • Introspection enabled
  • Authorization per field/type
  • Query depth limits
  • Query complexity limits
  • Batching vulnerabilities
  • Injection attacks
  • Information disclosure

How to Test

Step 1: Find GraphQL Endpoints

#!/bin/bash
TARGET="target.com"

endpoints=(
    "/graphql"
    "/graphql/console"
    "/graphiql"
    "/graphql/v1"
    "/api/graphql"
    "/gql"
    "/query"
)

for endpoint in "${endpoints[@]}"; do
    status=$(curl -s -o /dev/null -w "%{http_code}" \
        -X POST "https://$TARGET$endpoint" \
        -H "Content-Type: application/json" \
        -d '{"query":"{ __typename }"}')

    if [ "$status" != "404" ]; then
        echo "[FOUND] $endpoint: $status"
    fi
done

Step 2: Test Introspection

# Full introspection query
curl -s -X POST "https://target.com/graphql" \
    -H "Content-Type: application/json" \
    -d '{"query": "{ __schema { types { name fields { name type { name } } } } }"}' | jq '.'

# Get all queries and mutations
curl -s -X POST "https://target.com/graphql" \
    -H "Content-Type: application/json" \
    -d '{"query": "{ __schema { queryType { fields { name } } mutationType { fields { name } } } }"}' | jq '.'

Step 3: GraphQL Security Tester

#!/usr/bin/env python3
import requests
import json

class GraphQLTester:
    def __init__(self, url, headers=None):
        self.url = url
        self.session = requests.Session()
        if headers:
            self.session.headers.update(headers)
        self.findings = []

    def query(self, query, variables=None):
        """Execute GraphQL query"""
        payload = {"query": query}
        if variables:
            payload["variables"] = variables

        response = self.session.post(
            self.url,
            json=payload,
            headers={"Content-Type": "application/json"}
        )
        return response.json()

    def test_introspection(self):
        """Test if introspection is enabled"""
        print("[*] Testing introspection...")

        query = """
        {
            __schema {
                types { name }
                queryType { fields { name } }
                mutationType { fields { name } }
            }
        }
        """

        result = self.query(query)

        if 'errors' not in result or '__schema' in str(result):
            print("[VULN] Introspection enabled!")
            self.findings.append({
                "issue": "GraphQL introspection enabled",
                "severity": "Medium"
            })

            # Extract types
            if 'data' in result and result['data']:
                schema = result['data'].get('__schema', {})
                types = schema.get('types', [])
                print(f"  Found {len(types)} types")

                # List queries and mutations
                query_type = schema.get('queryType', {})
                if query_type:
                    queries = [f['name'] for f in query_type.get('fields', [])]
                    print(f"  Queries: {queries[:10]}")

        else:
            print("[OK] Introspection disabled")

    def test_authorization(self, queries):
        """Test authorization on queries"""
        print("\n[*] Testing authorization...")

        for query_name, query in queries.items():
            result = self.query(query)

            if 'data' in result and result['data']:
                print(f"[POTENTIAL] Accessible: {query_name}")
            elif 'errors' in result:
                error_msg = str(result['errors'])
                if 'unauthorized' in error_msg.lower() or 'forbidden' in error_msg.lower():
                    print(f"[OK] Protected: {query_name}")
                else:
                    print(f"[CHECK] Error on {query_name}: {error_msg[:50]}")

    def test_depth_limit(self, max_depth=10):
        """Test query depth limits"""
        print("\n[*] Testing depth limits...")

        # Build nested query
        nested = "{ users "
        for i in range(max_depth):
            nested += "{ friends "
        nested += "{ id } " + "} " * max_depth + "}"

        result = self.query(nested)

        if 'errors' in result:
            error = str(result['errors'])
            if 'depth' in error.lower() or 'complexity' in error.lower():
                print(f"[OK] Depth limit enforced")
            else:
                print(f"[WARN] Query failed but not due to depth limit")
        else:
            print(f"[VULN] No depth limit (tested {max_depth} levels)")
            self.findings.append({
                "issue": "No query depth limit",
                "severity": "Medium"
            })

    def test_batching(self):
        """Test batch query vulnerabilities"""
        print("\n[*] Testing batching...")

        # Batch login attempts (for brute force)
        batch_query = """
        query {
            q1: login(email: "[email protected]", password: "pass1") { token }
            q2: login(email: "[email protected]", password: "pass2") { token }
            q3: login(email: "[email protected]", password: "pass3") { token }
        }
        """

        result = self.query(batch_query)

        if 'errors' not in result or 'data' in result:
            print("[WARN] Batching allowed - potential for brute force")
            self.findings.append({
                "issue": "Batching allows brute force",
                "severity": "Medium"
            })

    def test_injection(self):
        """Test for injection vulnerabilities"""
        print("\n[*] Testing injection...")

        payloads = [
            '{ user(id: "1 OR 1=1") { id } }',
            '{ user(id: "1\'; DROP TABLE users--") { id } }',
            '{ user(id: "${7*7}") { id } }',
            '{ user(id: "{{7*7}}") { id } }',
        ]

        for payload in payloads:
            result = self.query(payload)
            if 'errors' in result:
                error = str(result['errors'])
                if 'sql' in error.lower() or 'syntax' in error.lower():
                    print(f"[VULN] SQL error disclosure")
                    self.findings.append({
                        "issue": "SQL injection possible",
                        "severity": "Critical"
                    })

    def generate_report(self):
        print("\n" + "="*50)
        print("GRAPHQL SECURITY REPORT")
        print("="*50)

        if not self.findings:
            print("\nNo issues found.")
            return

        print(f"\nFindings: {len(self.findings)}")
        for f in self.findings:
            print(f"\n  [{f['severity']}] {f['issue']}")

# Usage
tester = GraphQLTester(
    "https://target.com/graphql",
    {"Authorization": "Bearer token"}
)

tester.test_introspection()
tester.test_authorization({
    "users": "{ users { id email } }",
    "admin": "{ admin { secrets } }",
})
tester.test_depth_limit()
tester.test_batching()
tester.test_injection()
tester.generate_report()

Tools

ToolDescription
GraphQL VoyagerSchema visualization
AltairGraphQL client
InQLBurp Suite extension
graphql-copSecurity auditor

Remediation

# Disable introspection in production
from graphene import Schema

schema = Schema(query=Query, mutation=Mutation)

# Production settings
if not DEBUG:
    schema.introspection = False

# Implement depth limiting
from graphene.validation import depth_limit_validator

schema.execute(query, validation_rules=[depth_limit_validator(5)])

Risk Assessment

FindingCVSSSeverity
Introspection enabled5.3Medium
No depth limit (DoS)5.3Medium
Authorization bypass7.5High
SQL injection9.8Critical

Checklist

[ ] GraphQL endpoint found
[ ] Introspection tested
[ ] Authorization checked
[ ] Depth limits tested
[ ] Batching tested
[ ] Injection tested
[ ] Findings documented