Harold

@BlueRock_io 4-Learn more at mcp-trust.com

1-Last week, it was publicized that AWS had a 13 hour outage as a result of an agent deleting an environment. Not all tools are created equal. This is precisely why @BlueRock_io developed context-aware guardrails. #infosec #ai #aisecurity #mcp #mcpsecurity

@BlueRock_io 3-We facilitate context-awareness by leveraging the MCP Trust Registry which has analyzed over 9,300 MCP servers, provides a comprehensive security analysis and classifies available tools along the lines of read, write, cost-impact, destructive, exec-capable and other classes.

@BlueRock_io 2-This capability allows organizations to define policies to control what classes of tools can be executed. Preventing destructive actions and/or enabling read-only mode can help you build safely and monitor for actions that may have adverse impact.

Gateways are not flawed. They’re operating at the correct layer for what they were designed to do. But agentic systems have a second layer — and that’s where the failures emerge. #AgenticObservability

Agentic creates a split path challenge for a lot enterprises -- developers want to go fast and build better and security needs visibility and guardrails to reduce risk. We're pleased to announce the launch of our Agentic Observability PaaS available today for free.

Unbounded network fetch and SSRF can leak data and keys. It can also be identified and controlled. Here’s David Onwukwe’s write-up: bluerock.io/post/mcp-furi-… #mcp #mcpsecurity #ai