About
I'm Sanjeev, a technical architect focused on AI and security.I built AI agents for real tasks — travel, payments — and kept hitting the same wall: no way to control what an agent can do. Every framework hands agents an API key with full access. No scoping, no audit trail, no kill switch.OAuth solved this for apps 15 years ago. But agents aren't apps — they spawn sub-agents, act autonomously, and chain actions across services. We need a new protocol for the agentic world.That's why I built Grantex — an open authorization protocol for AI agents. Scoped permissions, delegation chains, real-time revocation, and audit trails.Three SDKs, eight integrations, all Apache 2.0. My goal: make Grantex the standard way agents prove what they're allowed to do.
Links
Badges
