Provider-specific requirements
Use scrape or retrieval actions for one-off reads. Use crawl, extract, batch, monitor, and WebDB actions only when the broader or persistent workflow is intentional. Read the current resource before updating or deleting it. Confirm IDs before monitor runs, batch cancellation, WebDB sync or reprocessing, purges, and deletes. These operations can change customer-owned state. Start withcontextdev_get_web_scrape_markdown for a single page and contextdev_post_web_crawl for a bounded site crawl. Keep page limits and timeouts small while validating a workflow.
Find an applicable action
Confirm your effective workspace, then search the live catalog:Inspect the selected contract
Replace<returned-tool-id> with the identifier returned by discovery:
Prepare one reviewed input
Use one authorized record that matches the inspected schema. Preserve its source identity and define the result you will accept. Open the action reference for its payload and response shape. Provider-backed execution can incur usage or external effects. For reusable logic, follow Write Plays. For direct-call recovery, use Tool execution recovery.Continue only when the workspace, provider, required fields, acceptance rule, and applicable price are understood. Keep unavailable access separate from no-result evidence.