Live survey286 models · 13 providers · 17,592 auditsAudited Using Git Worktrees · SAFE →What changedHow we verifyllms.txt

MCP servers · publisher

MCP servers by agentmail-to

agentmail-to publishes 3 MCP servers in the atlas. 3 are SAFE, 0 CAUTION and 0 BLOCK. Each was read in full before it was listed.

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
CoCopilotAIGoogle AI Mode
Cite

Cite this page

ModelsAtlas. “MCP servers by agentmail-to”. Retrieved 28 Oct 2026. https://modelsatlas.com/mcp/agentmail-to

Verdicts · 3 servers

✓SAFE3!CAUTION0✕BLOCK0

Transports

  • stdio, runs locally2
  • Streamable HTTP, remote2
  • SSE, legacy remote0
  • Not declared1

A server can list more than one transport.

Tool risk

1

server expose at least one destructive tool: delete, overwrite or run. The median server with a readable tool list exposes 4 tools.

Servers with no destructive tools →

3 servers by agentmail-to

1–3 of 3 · ranked by trust score, then stars

AgentMail Toolkit

agentmail-to/agentmail-toolkit

✓SAFE B
★ 1074 destructive

agentmail-mcp

agentmail-to/agentmail-mcp

✓SAFE B

Email inboxes for AI agents: send, receive, reply, search, and manage threaded email over MCP.

★ 684 readstdio

AgentMail

agentmail-to/agentmail-mcp

✓SAFE B

Email inboxes for AI agents: send, receive, reply, search, and manage threaded email over MCP.

★ 663 readstdio

More from agentmail-to

agentmail-to as a publisher →

Questions

Which agentmail-to MCP server has the best trust score?
AgentMail Toolkit (agentmail-to/agentmail-toolkit) and agentmail-mcp (agentmail-to/agentmail-mcp), ranked by trust score and then stars.
Are agentmail-to's MCP servers read-only?
2 of 3 declare only read tools; 1 expose at least one destructive tool; 0 build their tool list at runtime, so we cannot read it from the source.

Keep going in your AI

Opens with this page as the source. We add nothing else to the prompt.