Centaur

Use Centaur's MCP tools for actions and context exposed by the user's deployment.

Workflow

1. When identity or authorization matters, call centaur_whoami before other Centaur tools.
2. If the centaur MCP tool is listed, use it to list or search the tool catalog when the correct tool is unclear.
3. When centaur is available, inspect an unfamiliar CLI progressively with {"command": ["run", "<tool>", "--help"]}. Add known command and subcommand segments before --help until the relevant options and arguments are shown. Treat that help output as the current contract.
4. When centaur is available, run the command with {"command": ["run", "<tool>", "<argv>", "..."]}. Pass every CLI argument as a separate token and do not guess command names or options.
5. If centaur is not listed or cannot complete the request, use the legacy per-service MCP tools. Each accepts a method and arguments object. Call it with method: "help" when its methods or arguments are unclear.
6. Summarize consequential writes and return relevant identifiers or links.

Centaur authorizes calls using the signed-in principal's live roles and grants. Never request, paste, print, or store Centaur OAuth tokens.

Connection recovery

If no Centaur tools are available, explain that the client still needs the deployment-specific MCP endpoint.

• Codex: register it with codex mcp add centaur --url <CENTAUR_MCP_URL>, then run codex mcp login centaur. The --url flag is required for Streamable HTTP and OAuth.
• Claude Code: configure the plugin's mcp_url, open /mcp, and authenticate the centaur server.
• Other MCP clients: configure a remote HTTP server named centaur with the deployment's /mcp URL and complete its OAuth flow.

Do not substitute a guessed hostname. Ask for the deployment URL when it is not already configured or supplied.