@canner-ca/mcp is a Model Context Protocol server that exposes Canner as native tools for an AI agent — Claude Desktop, Cursor, and others. Zero dependencies; it speaks the MCP stdio transport directly. Marketing overview on the MCP page.

Setup

You need a Canner API token (cnr_…) from Dashboard → Account. Add the server to your MCP client’s config — claude_desktop_config.json for Claude Desktop, .cursor/mcp.json for Cursor — then restart the client and the Canner tools appear.

{
  "mcpServers": {
    "canner": {
      "command": "npx",
      "args": ["-y", "@canner-ca/mcp"],
      "env": { "CANNER_TOKEN": "cnr_your_token_here" }
    }
  }
}

Already ran canner login with the CLI? Drop the CANNER_TOKEN line — the server falls back to the credentials stored at ~/.canner/credentials.

Tools

Each tool wraps one endpoint of the versioned /v1 REST API:

`canner_whoami`	Confirm which account the agent is signed in as.
`canner_list_projects`	List the projects this account can access.
`canner_get_project`	Get one project's status, URL, and source.
`canner_list_deployments`	List a project's deployments, newest first.
`canner_get_deployment`	Read one deployment in full — including its build log.
`canner_deploy`	Rebuild a GitHub-connected project from HEAD.
`canner_list_env_vars`	List a project's environment variables.
`canner_set_env_var`	Add an environment variable.
`canner_list_domains`	List a project's custom domains.
`canner_add_domain`	Attach a custom domain.
`canner_verify_domain`	Re-check DNS and provision TLS for a domain.

Safe by default

Destructive operations — deleting a project, revoking a token, dropping a database or domain — are deliberately notexposed, so an agent can’t take them by mistake. Do those from the dashboard or the CLI.

MCP server.

Setup

Tools

Safe by default

Related