Why AI Agent Sandboxes Aren't Firewalls: Moving Beyond Post-Action Observability to In-Line PreToolUse Interdiction > ThumbGate Engineering Daily | 2026-10-09 > Author: Igor Ganapolsky @IgorGanapolskyhttps://github.com/IgorGanapolsky > Verified...

Why AI Agent Sandboxes Aren't Firewalls: Moving Beyond Post-Action Observability to In-Line PreToolUse Interdiction > ThumbGate Engineering Daily | 2026-10-09 > Author: Igor Ganapolsky @IgorGanapolskyhttps://github.com/IgorGanapolsky > Verified...

Why AI Agent Sandboxes Aren't Firewalls: Moving Beyond Post-Action Observability to In-Line PreToolUse Interdiction > ThumbGate Engineering Daily | 2026-10-09 > Author: Igor Ganapolsky @IgorGanapolskyhttps://github.com/IgorGanapolsky > Verified...

A person should see the calls that change something. Reads can run when a policy already allows that tool and those arguments. The failure mode Reviewing every tool call does not scale, and reviewing none of them leaves a coding agent free to write...

A system prompt asks the model to behave. The tool call still goes out when the model ignores that ask. ThumbGate sits in front of the call. The PreToolUse hook reads the command before it runs and can block it. A thumbs-down on a bad call becomes a...

Every week, a new paper or blog post claims to have found the "magic system prompt" that prevents LLM prompt injection. And every week, someone with a creative base64 payload, zero-width spaces, or a nested translation trick shatters it in five minut...