Claude Code & Other Agents

Command isn’t limited to the built-in chat interface. Any MCP-capable agent can connect with the same tool set that Command Center uses in the browser — this is the operator tier outside the app, primarily for the cegtec team during day-to-day operations, but open to anyone who wants to go deeper.

Connecting

The MCP endpoint has the form https://<domain>/api/mcp/<key>. Find the key under Extensions in the web app, or retrieve it via get_mcp_key (or generate_mcp_key to rotate it) from an already-connected session. Add this endpoint as an MCP server in your agent — nothing more is required.

MCP is an open standard, not a cegtec-proprietary protocol: Claude Code is the reference client, but Codex CLI, OpenClaw, Cursor, Claude Desktop, and any other MCP client connect the same way and see the same tools.

What the agent gets on connect

Every client receives the server instructions on connect — the InitializeResult.instructions part of the MCP protocol — explaining the system model: the nine terms, the object model, the rules every agent must follow. The explain_system tool returns the same orientation again at any time, as structured JSON. Both are standard MCP mechanics, so they work identically for every client, not just Claude Code.

More than the customer tier

Operator access sees the full tool set, not just the curated set in Command Center — including bulk operations across many rows and playbooks at once, and the ability to run entire workflows as a script instead of one chat message at a time. The full generated reference of every tool, by category, lives in the MCP Tool Reference.

Governance still applies

Approvals, workspace isolation, and the credit cap are enforced server-side, not client-side — they apply no matter which agent connects, because they’re bound to the MCP key, not the client software. Even an operator agent proposes; a human approves. See Approvals and Costs.

See also: Command for the customer tier of the same interface.