Forem

Andrea profile picture

Andrea

Co-founder @ Sentinel Gate — open-source security for MCP. Building the missing guardrails for AI agents.

Location London Joined Joined on  Personal website https://github.com/Sentinel-Gate/Sentinelgate
MCP security has 4 layers. Most teams have 2.
Cover image for MCP security has 4 layers. Most teams have 2.

MCP security has 4 layers. Most teams have 2.

1
Comments
4 min read

Want to connect with Andrea?

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

Already have an account? Sign in
I let my AI agent read a file. It tried to leak my credentials.
Cover image for I let my AI agent read a file. It tried to leak my credentials.

I let my AI agent read a file. It tried to leak my credentials.

1
Comments
4 min read
Your AI agent sandbox has no gate
Cover image for Your AI agent sandbox has no gate

Your AI agent sandbox has no gate

1
Comments
5 min read
Stop your AI agent from writing files it shouldn't — in under a minute
Cover image for Stop your AI agent from writing files it shouldn't — in under a minute

Stop your AI agent from writing files it shouldn't — in under a minute

1
Comments 2
2 min read
We kept thinking SentinelGate was ready. It wasn't.
Cover image for We kept thinking SentinelGate was ready. It wasn't.

We kept thinking SentinelGate was ready. It wasn't.

Comments
6 min read
What's missing from the --dangerously-skip-permissions safety playbook
Cover image for What's missing from the --dangerously-skip-permissions safety playbook

What's missing from the --dangerously-skip-permissions safety playbook

Comments
6 min read
Your agent doesn't need one security tool that does everything. It never did.
Cover image for Your agent doesn't need one security tool that does everything. It never did.

Your agent doesn't need one security tool that does everything. It never did.

Comments
6 min read
An AI safety researcher's agent deleted her inbox. The fix isn't a better prompt.
Cover image for An AI safety researcher's agent deleted her inbox. The fix isn't a better prompt.

An AI safety researcher's agent deleted her inbox. The fix isn't a better prompt.

1
Comments
6 min read
We built a firewall for AI agents. It doesn't use AI.
Cover image for We built a firewall for AI agents. It doesn't use AI.

We built a firewall for AI agents. It doesn't use AI.

Comments
7 min read
What the OpenClaw and Moltbook Breaches Reveal About AI Agent Security

What the OpenClaw and Moltbook Breaches Reveal About AI Agent Security

Comments
9 min read
Your MCP agents have no guardrails. Here's how to fix that.
Cover image for Your MCP agents have no guardrails. Here's how to fix that.

Your MCP agents have no guardrails. Here's how to fix that.

Comments
2 min read
Your MCP agents have no guardrails. Here's how to fix that.
Cover image for Your MCP agents have no guardrails. Here's how to fix that.

Your MCP agents have no guardrails. Here's how to fix that.

Comments
2 min read
loading...