Invicti Pentest Tool Benchmarks 52 Unique Vulnerabilities

The hybrid AI security platform completes target assessments in under an hour while linking findings directly to source files.

Updated on Oct. 25, 2026 in Cybersecurity

Bold flat-color editorial illustration showing a sculptural, interlocking geometric structure in navy, cream, and gold, representing autonomous cybersecurity architecture. — Invicti released benchmark results for its new agentic pentest platform, which successfully identified 52 vulnerabilities during recent hybrid security assessments. AI Illustration. Upload story photo >

Live Poll

Do you trust automated security tools to reliably detect and report vulnerabilities in your software?

Invicti has published benchmark results for its Agentic Pentest tool, documenting the identification of 52 unique vulnerabilities across four security assessments. This research-stage performance data highlights a hybrid testing architecture that completed assessments of Fider and Photoview in under an hour each.

Why it matters

The shift toward autonomous security testing seeks to balance automated speed with the depth of human-led penetration testing. By comparing hybrid agentic results against existing DAST benchmarks, security teams can better quantify the trade-offs between rapid scanning and deeper, autonomous investigative techniques.

Invicti completed assessment of Fider in 46 minutes and Photoview in 50 minutes, identifying 27 vulnerabilities via DAST and 25 via agentic testing. The report includes 31 direct links to source files.

The players

Invicti

A cybersecurity firm specializing in automated web application security testing and DAST solutions.

The details

The system utilizes a hybrid architecture that integrates Dynamic Application Security Testing (DAST) — a process that examines applications while they are running to find security holes — with autonomous AI-driven investigation. This approach allows the agentic tools to perform both authenticated black-box tests and source-assisted assessments. By combining these methods, the platform attempts to correlate runtime vulnerabilities with specific lines of code.

September 27-28, 2026: Benchmark assessments were conducted by the research team.
October 24, 2026: Invicti published the final benchmark results.
December 31, 2026: Introductory pricing for these security assessments concludes.

The Tech Race

Invicti is positioning its agentic pentest platform against established security competitors like Aikido and XBOW. The results represent an attempt to standardize how security teams measure the effectiveness of AI-driven vulnerability discovery compared to traditional scanning.

Security engineers and developers can access this pentesting capability immediately, with introductory pricing in effect through the end of the year. The platform is designed to streamline vulnerability identification for web applications like Fider and Photoview.

The takeaway

The benchmark suggests that hybrid agentic tools can maintain high-speed testing while providing granular links to source code. Industry practitioners should watch for subsequent third-party validation or additional public benchmarks that include direct performance data for competitors like Aikido and XBOW.