SERVICE & TOOLS

Octoparse Collection

Official Octoparse (Bazhuayu) cloud-collection MCP — search templates, run tasks and export structured data (OAuth required). Collection entry, not DataHub

devOAuth

6 available tools

get_task_status

Check one Octoparse task by taskId. Returns its normalized unexecuted/running/completed/stopped status, current lotNo when available, and collectedRows. collectedRows is a non-negative integer that de…

search_tasks

Search the current user's existing Octoparse tasks. Use this to find a taskId before export_data or start_or_stop_task. Supports pagination, keyword, status, and explicit taskIds. When UI rows are sho…

search_templates

Use this first for scrape, crawl, extract, or collect requests. Choose one selector: `query` for semantic discovery or `id`/`slug` for exact lookup; `limit` works with `query` only (1-20). `query` ret…

execute_task

Start an Octoparse cloud task from `templateName` + `parameters` (JSON object string keyed by `inputSchema[].field`). Preflight runs automatically and returns actionable input errors without creating …

export_data

Read one ClientAPI Storage page for an existing Octoparse lot. Octoparse taskId and lotNo are required. Defaults are page=1 and pageSize=20; maximum pageSize is 100. Result data is the current page Bo…

start_or_stop_task

Start or stop an existing Octoparse task by taskId. Use action=`start` or `stop`.