Product releases, security research, company announcements, and events from the Infiltra offensive AI engineering team.
We proudly unveiled our cutting-edge autonomous security solution to the APAC community in Melbourne. Discover how we're redefining automated vulnerability testing with deterministic proof-of-exploit validation, real-time agent observability, and verified execution traces.
Autonomous web pentesting requires authenticated access. Learn how our browser agent handles TOTP generation automatically and coordinates human-in-the-loop workflows for push and SMS MFA before injecting valid state into security scans.
A technical deep-dive into prompt injection propagation across MCP servers, poisoned tool manifests, and indirect context hijacking in multi-agent environments with verified exploit chains.
Strengthening regional compliance: 100% Australian data residency is now live for all agent sandboxes, scanning telemetry, and vulnerability reports, meeting APRA CPS 234 and Privacy Act requirements.
How our multi-agent validation loop writes, executes, and records harmless reproduction traces before flagging an alert, eliminating triage noise and giving engineering teams reproducible steps.
Our offensive security research team breaks down the industry shift from static rule dictionaries to autonomous reasoning agents that replicate human penetration testing workflows.
Want to see our autonomous pentesting and red-teaming agents in action on your own applications and LLMs?