Test resources

Staging routes, test accounts, and sample data left in live deployments, often with weaker access rules than production features.

1분 분량2026년 10월 2일 업데이트florianamette기여자 1명
어시스턴트에게 묻기 (MCP)GitHub에서 편집

Context#

“Test” and “demo” features are often implemented with relaxed rate limits, known passwords, or extra debug fields. They can share the same code path as production features with a feature flag, which makes them high-yield in scope when the flag is wrong in prod.

Theory#

Discovery comes from static strings in client JS, robots.txt, DNS names (staging.example.com pointing to prod-like infra), and error messages that name test tenants. The abuse is usually the same BOLA or function-level class with a smaller guard surface.

Practice#

Search client artifacts for test hostnames and flags#

  • Extract strings from mobile and web bundles for test, demo, internal subdomains, then try those hostnames and path prefixes on the in-scope program with the same test accounts you already have.

Tools#

  • strings
  • Burp Suite
  • curl