Copy-paste prompts

{
  "mcpServers": {
    "lumen-crawler-docs": {
      "url": "https://crawler-docs.lumen-labs.ai/docs/mcp"
    }
  }
}

{
  "mcpServers": {
    "lumen-crawler-docs": {
      "url": "http://localhost:3001/docs/mcp"
    }
  }
}

{
  "mcpServers": {
    "lumen-crawler-docs": {
      "command": "npx",
      "args": ["-y", "@pyroprompts/mcp-stdio-to-streamable-http-adapter"],
      "env": {
        "URI": "https://crawler-docs.lumen-labs.ai/docs/mcp",
        "MCP_NAME": "lumen-crawler-docs"
      }
    }
  }
}

You are integrating Lumen Crawler, a Playwright crawl task API.

Docs: https://crawler-docs.lumen-labs.ai/docs/llms.txt
OpenAPI: https://crawler-docs.lumen-labs.ai/docs/openapi.json
API: https://crawler.lumen-labs.ai
Console: https://crawlapp.lumen-labs.ai
Docs MCP: search_docs, list_sections, get_page, get_code_example

Rules:
- Authenticate crawl calls with Authorization: Bearer <lc_live_ key from the console>. The secret is shown once.
- POST /v1/crawls with a unique Idempotency-Key and JSON {"url":"...","max_pages":N}. Expect HTTP 202 and an id.
- Poll GET /v1/crawls/{id} until status is completed, cancelled, or failed. There are no webhooks. 404 means unknown crawl, not pending.
- Read Markdown with GET /v1/crawls/{id}/pages?include=markdown. Paginate using cursor = last item urlHash.
- GET /map is in-universe only. GET /doors lists off-universe URLs and never includes bodies. To crawl another host, start a new crawl. follow_doors: true returns 422.
- GET /cleaned is a second wait after crawl status. 409 cleanup_not_ready means wait. 404 means cleanup was skipped.
- On 402 quota_exceeded, max_pages does not fit remaining included pages. Do not retry the same max_pages. Failed fetches and robots skips do not consume quota; pages_ok does.
- On 401 unregistered_key, the Bearer token is not a registered console key.
- Production is HTTPS-only. Private IPs and metadata hosts are blocked.
- Results expire after 30 days. User-agent token for robots.txt is LumenCrawler.