# Cloro — MCP server on mcp.ai
> Connect your Cloro account and use 10 tools for web scraping straight from your AI agent. Connect with your own API key. Cloro provides structured data from Google Search and leading AI assistants, with synchronous extraction and asynchronous task APIs.

By: mcp.ai · official
Page: https://mcp.ai/cloro

## Connect (MCP protocol)
Remote MCP endpoint (HTTP, streamable): https://api.mcp.ai/p_cloro?ms=1787293620000
Add it as a custom/remote MCP connector, then authenticate when prompted.

## REST API (no MCP client required)
Every tool is also a REST endpoint, authed with a workspace API key.
Discover: GET https://api.mcp.ai/api/cloro/_endpoints   # public; lists every endpoint
Call:     POST https://api.mcp.ai/api/cloro/<endpoint>
  Authorization: Bearer sk_live_…   # create one at https://mcp.ai/settings/api-keys
  Content-Type: application/json
  Body: { …args }   →   { "ok": true, "tool": "<id>", "result": { … } }

## Developer docs
How to use (MCP or REST), markdown: https://mcp.ai/cloro/skill.md
Postman collection (v2.1): https://mcp.ai/cloro/postman.json

## Tools
- cloro_extract_ai_answer(state?: string, prompt: string, country: string, provider: string, include_html?: boolean, include_markdown?: boolean, include_raw_response?: boolean) — Synchronously extract a Gemini or Perplexity answer for a prompt and country. This spends credits and includes Cloro's 2-credit synchronous surcharge. Use Submit Async Task for ChatGPT or Copilot, whi
- cloro_extract_google_ai_mode_answer(uule?: string, device?: string, prompt: string, country: string, location?: string, include_html?: boolean, include_markdown?: boolean) — Extract a structured Google AI Mode answer with sources and optional rich modules for a prompt and country. This synchronous call costs at least 6 credits.
- cloro_get_async_queue_status() — Read organization-wide queued and processing counts, priority distribution, and current async concurrency without modifying the queue.
- cloro_get_async_task(task_id: string, wait_for_completion?: boolean, poll_timeout_seconds?: number, poll_interval_seconds?: number) — Get a queued Cloro task by ID, or wait with bounded polling until it reaches COMPLETED or FAILED. FAILED is returned as a valid terminal record with its provider response details intact.
- cloro_get_credits() — Read the organization's current completed-charge credit balance and billing cycle. Queued async work is not reserved, and even after a task reports a completed charge, the balance may take about 25 se
- cloro_list_supported_countries(model?: string) — Return Cloro's current supported country codes, optionally filtered by provider model. This reports geographic support only and does not confirm that a provider monitor is operational.
- cloro_list_us_states() — Return the US state and DC codes accepted for state-level targeting. State targeting applies only to ChatGPT, Gemini, Copilot, and Perplexity and adds 2 credits.
- cloro_search_google(url?: string, uule?: string, pages?: integer, query?: string, device?: string, country?: string, location?: string, search_type?: string, include_html?: boolean, include_ai_overview?: boolean, include_paa_ai_overview?: boolean) — Extract structured Google web or Google News results from a query and country, or web results from a complete Google Search URL. This synchronous call costs at least 5 credits; extra result pages cost
- cloro_submit_async_task(task: object) — Queue one non-cancelable Cloro monitor task and return its task ID and estimated credit charge without waiting. An accepted task continues after this call and spends the reported credits on completion
- cloro_submit_async_task_batch(tasks: object[]) — Queue 1-500 independently validated, non-cancelable Cloro monitor tasks. Each accepted task continues after this call and spends its reported credits on completion. Inspect summary, partial_success, a

## Example prompts
- "What can I do in Cloro?"
- "Show me a summary of my Cloro account"

## Links
Docs: https://mcp.ai/docs/mcps/cloro
Website: https://mcp.ai/mcps/cloro