Products & research
Products, research and open-source work around AI-agent security.
My work focuses on one problem: making AI-agent work safe enough to adopt. Oktsec brings policy, control and evidence into agent environments. The open-source side maps risk across MCP servers, CLIs, automation and public agent tooling.
- Security platform · Policy · EvidenceOktsec
The security platform for AI-agent work. Oktsec helps teams define policy, govern approved agent environments and verify the evidence that comes back.
Visit - Open source · Agent workflowsAguara + nanostack
Open-source tooling for scanning, building and reviewing agent workflows. Aguara provides deterministic checks for agent and MCP risk; nanostack structures AI-assisted engineering work.
Visit - Ecosystem monitoring · MCP · Supply chainSignal
Security signal for the agent ecosystem. Signal maps risk across public agent tooling, MCP servers, CLIs, automation paths and repositories, starting with GitHub.
Visit