validating-cors-policies

Validating CORS Policies

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "validating-cors-policies" with this command: npx skills add jeremylongshore/claude-code-plugins-plus-skills/jeremylongshore-claude-code-plugins-plus-skills-validating-cors-policies

Validating CORS Policies

Overview

Validate Cross-Origin Resource Sharing configurations in web applications and APIs for security misconfigurations that enable unauthorized cross-origin access. This skill analyzes CORS headers, middleware configurations, and server response behavior to detect wildcard origins, reflected origins, credential leakage, and overly permissive header/method exposure.

Prerequisites

  • Access to the target codebase and configuration files in ${CLAUDE_SKILL_DIR}/

  • For live endpoint testing: WebFetch tool available and target URLs accessible

  • Familiarity with the web framework in use (Express, Django, Flask, Spring, ASP.NET, etc.)

  • Reference: ${CLAUDE_SKILL_DIR}/references/README.md for CORS specification details, common vulnerability patterns, and example policies

Instructions

  • Locate all CORS configuration points by scanning for Access-Control-Allow-Origin , cors() middleware, @CrossOrigin annotations, CORS policy builders, and server config directives (nginx add_header , Apache Header set ) using Grep.

  • Check for wildcard origin (Access-Control-Allow-Origin: * ) -- flag as severity high when combined with Access-Control-Allow-Credentials: true , which browsers reject but indicates a misunderstanding of the security model.

  • Detect origin reflection patterns where the server echoes back the Origin request header without validation -- search for code that reads the Origin header and sets it directly in the response. Flag as CWE-942 (Permissive Cross-domain Policy), severity critical.

  • Validate the origin allowlist: check that allowed origins use exact string matching rather than substring or regex patterns vulnerable to bypass (e.g., example.com.evil.com matching a check for example.com ).

  • Assess Access-Control-Allow-Methods -- flag if dangerous methods (PUT , DELETE , PATCH ) are exposed without necessity. Verify that preflight (OPTIONS ) responses include appropriate method restrictions.

  • Evaluate Access-Control-Allow-Headers -- flag wildcard header allowance or exposure of sensitive headers like Authorization , Cookie , or custom auth headers to broader origins than necessary.

  • Check Access-Control-Expose-Headers for leakage of internal headers (e.g., X-Request-Id , X-Internal-Trace ) to cross-origin consumers.

  • Verify Access-Control-Max-Age is set to a reasonable value (600-86400 seconds) to balance security with performance -- missing or excessively long max-age values deserve a low-severity note.

  • For live endpoints, issue preflight requests via WebFetch with various Origin values (legitimate, malicious, null) and analyze the response headers to confirm server behavior matches the codebase configuration.

  • Compile findings with severity ratings, map to OWASP Testing Guide OTG-CLIENT-007, and provide remediation with correct CORS middleware configuration examples.

Output

  • CORS configuration inventory: Table of all CORS-enabled endpoints, their allowed origins, methods, headers, and credentials settings

  • Findings report: Each finding includes severity, affected endpoint/file, CWE reference (CWE-942, CWE-346), observed behavior, and remediation code

  • Preflight test results: For live endpoints, a table of Origin values tested and the corresponding server responses

  • Remediation examples: Framework-specific CORS configuration snippets (Express cors() , Django django-cors-headers , Spring @CrossOrigin , nginx headers)

Error Handling

Error Cause Solution

No CORS configuration found CORS handled at infrastructure layer (CDN, API gateway) Check CDN/gateway configs (Cloudflare, AWS API Gateway, nginx) for CORS header injection

WebFetch blocked or timed out Target endpoint unreachable or rate-limited Verify URL accessibility; fall back to static codebase analysis of CORS middleware configuration

Inconsistent CORS behavior across endpoints Multiple CORS configurations at different layers Map each layer (application, reverse proxy, CDN) and document the effective policy per endpoint

Origin reflection false positive Dynamic origin validation with a secure allowlist Verify the allowlist logic uses exact matching; mark as informational if the implementation is secure

Preflight not triggering Request classified as "simple request" by the browser Note that simple GET/POST requests bypass preflight; test with custom headers to force preflight

Examples

Express.js CORS Middleware Audit

Scan ${CLAUDE_SKILL_DIR}/src/app.js and ${CLAUDE_SKILL_DIR}/src/middleware/ for cors()

configuration. Flag origin: true (reflects any origin) as CWE-942, severity critical. Recommend replacing with an explicit allowlist: origin: ['https://app.example.com', 'https://admin.example.com'] .

Nginx CORS Header Review

Grep ${CLAUDE_SKILL_DIR}/nginx/ for add_header Access-Control-Allow-Origin . Flag any $http_origin variable usage that reflects the origin without validation. Verify that Access-Control-Allow-Credentials is only set for origins in the allowlist using an if block or map directive.

API Gateway CORS Configuration

Review ${CLAUDE_SKILL_DIR}/infra/api-gateway.yaml or equivalent IaC definitions for CORS settings. Flag wildcard * in allowed origins when credentials are enabled. Verify that Access-Control-Allow-Methods is scoped to only the HTTP methods each endpoint actually supports.

Resources

  • MDN: Cross-Origin Resource Sharing

  • OWASP Testing for CORS (OTG-CLIENT-007)

  • CWE-942: Permissive Cross-domain Policy

  • CWE-346: Origin Validation Error

  • Fetch Standard: CORS Protocol

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Coding

backtesting-trading-strategies

No summary provided by upstream source.

Repository SourceNeeds Review
Coding

svg-icon-generator

No summary provided by upstream source.

Repository SourceNeeds Review
Coding

performance-lighthouse-runner

No summary provided by upstream source.

Repository SourceNeeds Review