What it does
agentglass checks a public web page for hidden instructions written to target AI agents, which is the pattern known as prompt injection. The publisher describes it as a scanner that accepts any public URL.
The resource fits cases where an AI agent will read content from public web pages and someone wants to check that content first. The publisher describes it as free and requiring no authentication.
Limitations and requirements
Scanning is limited to public URLs.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
net.agentglass/agentglass
Scan any public URL for hidden instructions aimed at AI agents (prompt injection). Free, no auth.
v0.1.0As shown on registry.modelcontextprotocol.io when we last read it