Security review

Is agent-toolkit-mcp-server MCP safe to give an agent?

destructive capableunknowndestructive capablehealthy

A factual risk summary built from agent-toolkit-mcp-server’s real tool surface, execution model, and verification history — not a vibe. Trust score 32/100.

01What it can do

Has tools that can delete or irreversibly change data.

62 tools observeddestructive presentwrite present

02Execution model

Transport not yet confirmed, so the execution model is unknown. Treat as untrusted until verified.

Installs via uvx (`uvx agent-toolkit-mcp-server`) — it pulls and executes third-party code; pin versions and review the source.

03Permissions & auth

Read & write files · destructive

The write/destructive access this server can exercise, inferred from its verified tools.

04Verification

handshakenot confirmed

runshandshake:failed · metadata:passed · handshake:failed · metadata:passed · handshake:failed · metadata:passed · handshake:failed · metadata:passed · metadata:passed · handshake:failed

last_checked2026-08-16 19:31Z

sourcesPyPI [p4]

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