Skip to content

FLAGSHIP CASE STUDY

Security engineering work, explained.

One current flagship deep dive into MCP tool-call security: the risk, the design constraints, what shipped, and how the outcome was verified.

Current flagship

mcpgate

mcpgate v1.1.0: Securing AI Agent Tool Calls

How I shipped reverse-channel prompt-injection defenses for an MCP security gateway, caught an error-channel bypass in review, and released a safer v1.1.0.

AI SecurityMCPGoAgent Governance

Read the flagship case study ->

More deep dives will follow when they are worth reading.

I would rather keep this page focused than pad it with shallow write-ups. The current flagship is the best representative sample of my security, backend, and AI-agent governance work.