Reports & Resources

Research, benchmarks, and data from the KushoAI platform on API testing, security, and AI-native engineering.

Published
Report
State of API Security 2026
Security failure patterns from 1.4M test executions across 2,600+ organizations, mapped to the OWASP API Security Top 10. Auth failures, coverage gaps, supply chain risk, and the role of AI in closing them.
1.4M
Executions
2,616
Organizations
2026
Published
→
Report
State of Agentic API Testing 2026
Observed patterns from 1.4 million test executions across 2,616 organizations on how engineering teams generate tests, structure workflows, detect failures, and integrate AI into their testing pipelines.
1.4M
Executions
64,459
Test suites
2026
Published
→
Eval
APIEval-20
A task benchmark for evaluating AI agents on real-world API testing. 20 scenarios across 7 domains, measuring bug-finding capability from schema and payload alone — with a public dataset on Hugging Face.
20
Scenarios
7
Domains
98
Planted bugs
→