Waxell blog cover: GitHub AI Agent Prompt Injection — Claude, Gemini, Copilot, No CVE

Comment and Control: The GitHub AI Agent Attack That Three Vendors Hushed

Comment and Control: The GitHub AI Agent Attack That Three Vendors Hushed

Researcher Aonan Guan hijacked Claude Code, Gemini CLI, and Copilot Agent via PR titles and hidden HTML comments. All three paid bug bounties. None filed a CVE. Here's what that means for your agents.

Logan Kelly

Waxell blog cover: Agent Versioning as a Governance Problem

Agent Versioning Isn't a Deployment Problem. It's a Governance Problem.

Agent Versioning Isn't a Deployment Problem. It's a Governance Problem.

Rolling back agent code is easy. Rolling back agent behavior is something else. Here's why agent versioning is a governance requirement, not just an ops task.

Logan Kelly

600 Firewalls in 5 Weeks — The Human-in-the-Loop Lesson from the FortiGate AI Attack

600 Firewalls in 5 Weeks: What the FortiGate AI Attack Teaches Us About Human Oversight

600 Firewalls in 5 Weeks: What the FortiGate AI Attack Teaches Us About Human Oversight

An AI agent compromised 600+ firewalls across 55 countries in 5 weeks — without a human approving each command. Here's what enterprise teams building agents need to learn from it.

Logan Kelly

Waxell blog cover: AI Agent Token Budget Enforcement — Why Alerts Aren't Enough

The $47,000 Agent Loop: Why Token Budget Alerts Aren't Budget Enforcement

The $47,000 Agent Loop: Why Token Budget Alerts Aren't Budget Enforcement

A 4-agent system ran for 11 days and burned $47,000 — with full observability running. Here's why cost alerts aren't enforcement, and what is.

Logan Kelly

Waxell blog cover: CIS Prompt Injection Report 2026 — 340% Surge in Enterprise AI Attacks

340% and Climbing: What the CIS Prompt Injection Report Means for Enterprise AI Agents

340% and Climbing: What the CIS Prompt Injection Report Means for Enterprise AI Agents

The CIS published a major prompt injection threat report in April 2026. Documented attacks are up approximately 340% year-over-year. Two-thirds go undetected for 72+ hours. Here's what it means.

Logan Kelly

Waxell blog cover: Enterprise AI Agent Governance — The 12% Problem

96% of Enterprises Run AI Agents. Only 12% Can Govern Them.

96% of Enterprises Run AI Agents. Only 12% Can Govern Them.

96% of enterprises run AI agents — only 12% have centralized control. Here's the technical reality behind the governance gap, and what closing it actually requires.

Logan Kelly

Waxell

Waxell provides observability and governance for AI agents in production. Bring your own framework.

© 2026 Waxell. All rights reserved.

Patent Pending.

Waxell

Waxell provides observability and governance for AI agents in production. Bring your own framework.

© 2026 Waxell. All rights reserved.

Patent Pending.

Waxell

Waxell provides observability and governance for AI agents in production. Bring your own framework.

© 2026 Waxell. All rights reserved.

Patent Pending.