Technical writings on authorization, security, and systems

HOME
TAGS
ARCHIVES
ABOUT
CONTACT
CATEGORIES
AI AGENT SECURITY
AUTHORIZATION & CAPABILITY SECURITY

Home Tags

Tags

Tags

capability-security4

prompt-injection1

Recently Updated

The Model Wasn't Malicious, the Boundary Held Anyway
OAuth 2.0 Device Authorization Grant Security (RFC 8628)
Prompt Injection Is an Authorization Bug: Capability-Bounded AI Agents
Capability Boundaries Have No Memory: Preventing AI Agent Data Exfiltration
The Signature Verified, the Authority Widened Anyway: Capability Serialization Attacks

Trending Tags

authorization security ai-agents capability-security rust oauth prompt-injection

© 2026 Senthil Siva. Some rights reserved.

Using the Chirpy theme for Jekyll.

Trending Tags

authorization security ai-agents capability-security rust oauth prompt-injection

New content available