Connexagents MCP runs on your own machine — the client starts it, so there is no endpoint to ping. 510 installs a week from npm. Last commit 5 Nov 2026.
MCP server to file a capability gap with ConnexAgents. A human operator routes it to a specialist.
Today is the operative word: we check Connexagents MCP every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add connexagents-mcp -e 'CONNEXAGENTS_TOKEN=<CONNEXAGENTS_TOKEN>' -- npx -y connexagents-mcp
{
"mcpServers": {
"connexagents-mcp": {
"command": "npx",
"args": [
"-y",
"connexagents-mcp"
],
"env": {
"CONNEXAGENTS_TOKEN": "<CONNEXAGENTS_TOKEN>"
}
}
}
}
[mcp_servers.connexagents-mcp]
command = "npx"
args = ["-y", "connexagents-mcp"]
env = { CONNEXAGENTS_TOKEN = "<CONNEXAGENTS_TOKEN>" }
{
"mcpServers": {
"connexagents-mcp": {
"command": "npx",
"args": [
"-y",
"connexagents-mcp"
],
"env": {
"CONNEXAGENTS_TOKEN": "<CONNEXAGENTS_TOKEN>"
}
}
}
}
{
"servers": {
"connexagents-mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"connexagents-mcp"
],
"env": {
"CONNEXAGENTS_TOKEN": "<CONNEXAGENTS_TOKEN>"
}
}
}
}
Put your own value in place of
<CONNEXAGENTS_TOKEN> (ConnexAgents agent key (cxa_...)).
The author’s registry entry marks it as required.
Optional settings from the registry entry, left out of the block above:
CONNEXAGENTS_API_URL (API URL, ending in /api/v1), CONNEXAGENTS_AGENT_ID (UUID of the agent linked to this key. Required only for list_match_proposals).
Manage ipTIME routers through a local, safety-gated MCP server with firmware-aware capabilities.
MCP server for AI agents to dispatch real-world tasks to verified human operators
Contact point for AI agents: describe a missing capability or need; a human operator helps.
Live index of AI agents, MCP servers and tools with observed liveness and capability search.
Semantic and scratch 3D CAD for AI agents via CAiD + OCCT, with runtime capability discovery.
Security-first MCP server that connects any OpenAPI, GraphQL, gRPC or SOAP API to AI agents.
Human-in-the-loop for AI agents over MCP elicitation: confirm/form dialogs plus a capability doctor
Scan an MCP server or agent config for injection, exfiltration, and risky capabilities.
Answers built from our own checks of this server.