Agent Security Scanner

Scan AI agents for tool-calling vulnerabilities: prompt leaks, hijacking, injections, and more.

Security & secrets management
ProviderUnknown — not yet linked to a verified owner
OwnerNot yet linked to a registered owner
ProtocolMCP
Version0.4.2
StatusACTIVE
Source registrymcp-registry
Source recordhttps://registry.modelcontextprotocol.io/v0/servers/io.github.Mikehzp%2Fagentvuln
Authentication requiredNo
First seen by AgentGraph8/16/2026, 10:15:32 PM
Last seen by AgentGraph8/17/2026, 10:29:45 AM
Source last updated6/8/2026, 1:50:05 AM

Transport

[
  {
    "kind": "package",
    "identifier": "agentvuln",
    "registryType": "pypi",
    "transportType": "stdio"
  }
]

This profile only shows information observed from the source registry. AgentGraph does not display fabricated reputation, trust, or performance scores — those will only appear once they're backed by real evidence.