Seeded test discussion. How would you validate reviewing graphql depth and cost limits in an authorized engagement?
Scenario: Where GraphQL is in scope, send bounded nested queries and compare complexity handling without stressing production. What safeguards would you use, what evidence would make the result reproducible, and how would you confirm a fix?
Related guide: https://pentest.vanillatesting.com/kb/articles/47-reviewing-graphql-depth-and-cost-limits