DEV Community

Sofia_ Humanbound profile picture

Sofia_ Humanbound

Building Humanbound. An open-source adversarial testing engine, SDK, and CLI for AI agents. Join me in the issues https://github.com/humanbound

Agentic AI Security This Week: A Saturated Benchmark, 11 Framework CVEs, and 15 Competitors Who Finally Agree on Something

Agentic AI Security This Week: A Saturated Benchmark, 11 Framework CVEs, and 15 Competitors Who Finally Agree on Something

1
Comments
5 min read

Want to connect with Sofia_ Humanbound?

Create an account to connect with Sofia_ Humanbound. You can also sign in below to proceed if you already have an account.

Already have an account? Sign in
The Taiwan Attack: When an AI Agent Swarm Ran a Government Hack With No One Watching

The Taiwan Attack: When an AI Agent Swarm Ran a Government Hack With No One Watching

2
Comments
3 min read
When the model that finds the bug is the same model that could exploit it

When the model that finds the bug is the same model that could exploit it

1
Comments
6 min read
We're building a community library of agent attack scenarios, mapped to OWASP.

We're building a community library of agent attack scenarios, mapped to OWASP.

1
Comments
2 min read
The approval prompt was never the control. Black Hat just admitted it.

The approval prompt was never the control. Black Hat just admitted it.

2
Comments
5 min read
Trust Boundary Report, Issue 02: The Month Agentic AI Stopped Being a Thought Experiment

Trust Boundary Report, Issue 02: The Month Agentic AI Stopped Being a Thought Experiment

3
Comments 2
5 min read
Building a Public Backlog of AI Agent Failures: What's the Worst Thing Your Tests Didn't Catch?

Building a Public Backlog of AI Agent Failures: What's the Worst Thing Your Tests Didn't Catch?

10
Comments 4
2 min read
AI Security Means Two Different Things. Mythos Made That Visible.

AI Security Means Two Different Things. Mythos Made That Visible.

9
Comments 2
7 min read
A new paper argues that your prompt injection defence can't win.

A new paper argues that your prompt injection defence can't win.

7
Comments 1
2 min read
You're Still Alt-Tabbing to a Security Tool

You're Still Alt-Tabbing to a Security Tool

4
Comments
6 min read
We put adversarial agent testing directly in Claude Code and Cursor

We put adversarial agent testing directly in Claude Code and Cursor

10
Comments
3 min read
Beyond Moderation: Why LLM Systems Need a Policy Layer

Beyond Moderation: Why LLM Systems Need a Policy Layer

1
Comments
5 min read
Why Your AI Agent's Biggest Vulnerability Isn't a Missing Firewall

Why Your AI Agent's Biggest Vulnerability Isn't a Missing Firewall

7
Comments 2
8 min read
The Enforcement illusion: Why AI Agent Security Starts with Testing, Not Walls

The Enforcement illusion: Why AI Agent Security Starts with Testing, Not Walls

2
Comments 1
6 min read
AI agent evaluation is evolving. Here's what we're building.

AI agent evaluation is evolving. Here's what we're building.

7
Comments 1
1 min read
Taking a Proactive, Governance-Based Approach to API Security

Taking a Proactive, Governance-Based Approach to API Security

4
Comments
6 min read
Pulse Report - Key Insights from the 23rd Edition of the Developer Nation Survey

Pulse Report - Key Insights from the 23rd Edition of the Developer Nation Survey

1
Comments
1 min read
loading...