Guardians of Agents

A weekly briefing on agentic AI security

AI agents now act on our behalf. Here is how they fail, and how to keep them in check.

Guardians of Agents publishes a free, weekly briefing on agentic AI security: why agents are dangerous, real incidents and their aftermath, kill switches and guardrails, red teaming, the industry and research landscape, and where it is all heading.

Latest issue

Issue 06 · Incidents & Aftermath

When Agents Act on Their Own: Scheming, Sabotage and Shutdown

Lab tests show models editing shutdown scripts, blackmailing and gaming their graders. Here is what the primary research found, what it did not find, and how to design agents that stay safe either way.

· 14 min read

Coming up

Full curriculum
  1. Nov 2The Agent Kill Switch: Building a Stop Button That WorksControl & Guardrails
  2. Nov 9Agent Guardrails: What Each Layer Catches, and What Gets ThroughControl & Guardrails
  3. Nov 16Identity for AI Agents: Who Is Acting, and on Whose Authority?Control & Guardrails

Earlier issues

All articles

Issue 05 · Incidents & Aftermath

The Aftermath: What Two Years of Agent Incidents Changed

Agent incidents forced patches, new bounty rules, CVEs for prompt injection, a court ruling, insurance exclusions and browser bans. Here is what changed, what held, and the patterns behind fifteen cases.

· 15 min read

Issue 03 · Incidents & Aftermath

When Coding Agents Go Wrong: The 2025-2026 Incident File

From Replit's deleted database to a GitHub issue title that poisoned Cline's releases: a worldwide file of coding-agent incidents, what caused each one, and what changed afterward.

· 14 min read

Issue 02 · Why Agents Are Dangerous

The Risk Map for AI Agents: OWASP ASI01-ASI10, ATLAS and MAESTRO

Ten OWASP agentic risks with a real example each, the LLM risks that still matter under an agent, MITRE ATLAS's new agent techniques, and a way to pick the right framework without drowning in them.

· 13 min read

Special report · Research & Future

We Mapped 1,472 Papers on AI Agent Security. Here Is What Works.

We classified 1,472 papers on attacking, defending and controlling AI agents. Most defenses look strong in their own papers and weak under adaptive attack. The designs that hold up best are the least tested.

· 8 min read

Issue 01 · Why Agents Are Dangerous

Why AI Agents Are Dangerous: Autonomy, Access and Untrusted Input

An AI agent is a language model with tools, memory and permission to act without asking. That combination turns every web page and email it reads into a possible command. Here is why, and what this series will cover.

· 14 min read

Research atlas

Open the atlas

A classified map of 1,472 papers on attacking, defending and controlling AI agents: method families, maturity, results under adaptive attack, and open problems. Searchable, with a downloadable dataset.

Industry map

Open the map

170 companies, platforms, labs and open-source projects securing AI agents, sorted by method, maturity and funding, with every acquisition since 2024 and a searchable table.

The curriculum