Skip to main content
The cloro plugin for OpenClaw is a thin client over cloro’s sync monitor endpoints — the same POST /v1/monitor/* requests documented in the API reference — exposed as agent tools. Ask your OpenClaw agent to “check whether Google’s AI Overview for ‘best ai seo tools’ mentions us” and it calls the cloro API and reasons over the parsed JSON.

​
Prerequisites

​
Install the plugin

Install from the cloro OpenClaw plugin repository with the OpenClaw CLI:

​
Configure your API key

Set the key as an environment variable available to the OpenClaw gateway:
Or add it to openclaw.json under the plugin’s config block:
Optional config: baseUrl (defaults to https://api.cloro.dev), defaultCountry (used when a tool call omits country, defaults to us), and timeoutSeconds (defaults to 300 — monitor requests drive real provider sessions and can take a while).

​
Available tools

Assistant tools take a prompt with optional country and state (US state targeting). cloro_google additionally supports location, device, pages, and an aiOverview flag that includes Google’s AI Overview as markdown with sources. Each call consumes credits at the standard per-provider rates — see Providers and Billing & credits.

​
Alternative: connect the cloro MCP server

OpenClaw can also connect to cloro’s hosted MCP server directly for the same monitor tools. Pass your cloro API key in the URL path:
Or in openclaw.json:
The plugin remains the better fit for OpenClaw workflows — defaults and API key managed in one config block — but the MCP route works with any MCP client, and its guide lists the full tool set.

​
Example prompts

  • “Ask ChatGPT what the best SERP APIs are and list which vendors it cites.”
  • “Compare how Perplexity and Copilot answer ‘best CRM for startups’ — which sources do they cite?”
  • “Does Google’s AI Overview for ‘ai visibility tracking’ mention cloro? Check in the US and Germany.”
  • “Pull the latest Google News articles about our brand.”
Tool calls run synchronously and count against your concurrency limits. For large batches (tracking many prompts across engines), use the async API directly instead of looping tool calls.