agentservices-to / security
Security review
Is AgentServices MCP safe to give an agent?
destructive capableremotedestructive capablehealthy
A factual risk summary built from AgentServices’s real tool surface, execution model, and verification history — not a vibe. Trust score 33/100.
01What it can do
Has tools that can delete or irreversibly change data.
64 tools observeddestructive presentwrite present
02Execution model
Runs on the vendor's infrastructure; you connect over the network. No untrusted code runs on your machine, but you grant the hosted service access.
Connects to a remote URL — no local package execution.
03Permissions & auth
This server has write and destructive tools. No auth scopes are recorded for how that access is granted.
04Verification
handshakenot confirmed
runsmetadata:passed · handshake:failed · metadata:passed · handshake:failed · metadata:passed · handshake:failed · metadata:passed · handshake:failed · metadata:passed · handshake:failed
last_checked2026-10-13 16:21Z
sourcesUser submissions [p7]
Reduce the risk
Worried about handing an agent raw access? See governed agents in action — Apex gives your AI paced, capped, fully-logged hands with approval queues before anything runs.
Explore Apex →See also: full server page · setup · alternatives