Integration

Octoparse integration

Octoparse is a no-code web scraping platform. An agent can find a ready-made scraping template and run it as a new cloud task, start, stop and check existing tasks and their subtasks, read the rows a task collected (all rows, one run's rows, or only rows not yet exported), mark rows as exported or delete a task's data, duplicate or move tasks between task groups, read and update a task's URL and text-list loop items, and read per-task run and resource analytics. Most routes need a Standard plan; task editing and cloud runs need Professional or Enterprise, as each tool says. Not exposed: the username and password token routes, the URL-file upload route, and reading or listing Enter Text action values (they can hold website logins). Rate limit: 20 requests per second per the vendor, with stricter limits stated on single routes. Authentication is an Octoparse API key (starts with op_sk_) from the account center API keys page, sent in the x-api-key header.

Connect Octoparse How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Octoparse.Built from the 26 Octoparse tools below, on the connection you add once.

What you can do with Octoparse

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Task, 7 operations

duplicate_taskDuplicate taskPOSTCreates a copy of a task in a task group and returns the new taskId. Standard, Professional or Enterprise plan.
list_tasks_in_groupList tasks in groupGETLists the tasks (taskId, taskName) in one task group; taskGroupId comes from list_task_groups. Standard, Professional or Enterprise plan.
move_taskMove taskPOSTMoves a task to another task group. Returns moved true on success. Standard, Professional or Enterprise plan.
run_templateRun templatePOSTStep 2: creates and starts a real cloud scraping task from a template (from search_templates). Paid templates bill a fee per record collected: cap it with targetMaxRows. Returns status, taskId and lotNo for export_task_data.
search_tasksSearch tasksGETSearches your existing Octoparse tasks by keyword, status (Running, Stopped, Completed, Failed) or task IDs, with paging. Use the taskId with start_or_stop_task or export_task_data.
start_or_stop_taskStart or stop taskPOSTStarts or stops a cloud run of an existing task (taskId from search_tasks). Starting begins real data collection on your account; a task built from a paid template bills a fee per record collected. Then call export_task_data, following retryGuidance.
update_task_parametersUpdate task parametersPOSTChanges a task's saved configuration: loop item lists (URLs, text) and action properties, for the actions you send. Read actionId values with get_task_actions first. Professional or Enterprise plan.

Task data, 5 operations

delete_task_dataDelete task dataPOSTPermanently deletes ALL the data a task has collected. Cannot be undone. Standard, Professional or Enterprise plan.
get_non_exported_dataGet non exported dataGETReads rows of a task never marked as exported. After processing them call mark_data_exported, or the same rows come back on the next call. Standard, Professional or Enterprise plan.
get_run_dataGet run dataGETReads the rows collected by one run (lotno from start_cloud_run or run_template), starting at a row offset (0 for the first row). Standard, Professional or Enterprise plan.
get_task_dataGet task dataGETReads the rows a task collected, starting at a row offset (0 for the first row). Page with offset and size; restTotal says how many rows remain. Standard, Professional or Enterprise plan.
mark_data_exportedMark data exportedPOSTMarks the rows returned by get_non_exported_data as exported, so they stop appearing there. Standard, Professional or Enterprise plan.

Subtask, 3 operations

list_subtask_statusesList subtask statusesGETLists the subtasks (subTaskId, status) of the latest cloud run of a task, one page at a time. Professional or Enterprise plan.
start_subtasksStart subtasksPOSTStarts the given subtasks of a task in the cloud, which collects real data on your account (a paid template bills per record collected). subTaskIds come from list_subtask_statuses. Professional or Enterprise plan.
stop_subtasksStop subtasksPOSTStops the given subtasks of a task in the cloud. subTaskIds come from list_subtask_statuses. Professional or Enterprise plan.

Cloud run, 2 operations

start_cloud_runStart cloud runPOSTStarts a real cloud run of an existing task (a paid template bills per record collected) and returns its lotNo. Poll get_task_statuses_v2, then read rows with get_run_data. Professional or Enterprise plan.
stop_cloud_runStop cloud runPOSTStops the cloud run of a task. Professional or Enterprise plan.

Task action, 2 operations

get_task_actionsGet task actionsPOSTReads the Loop Item and Go to Web Page actions of tasks (their actionId, URLs and text lists), to edit with update_loop_items or update_action_properties. Professional or Enterprise plan.
update_action_propertiesUpdate action propertiesPOSTChanges the URL of a Go to Web Page action or the text of an Enter Text action in a task, for the actions you send. Professional or Enterprise plan.

Task status, 2 operations

get_task_statusesGet task statusesPOSTReads the cloud run status (taskId, taskName, status) of several tasks. Limited to 1 request every 5 seconds; prefer get_task_statuses_v2. Professional or Enterprise plan.
get_task_statuses_v2Get task statuses v2POSTReads detailed cloud run status of several tasks: status, rows extracted so far, run count, subtask count, start, end and next run times. Up to 5 requests per second. Professional or Enterprise plan.

Export, 1 operation

export_task_dataExport task dataGETStep 3: checks a task's export status (collecting, exporting, exported, no_data, failed, invalid) with up to 20 preview rows. Repeat after retryGuidance.retryAfterSeconds until exported, no_data, failed or invalid.

Loop item, 1 operation

update_loop_itemsUpdate loop itemsPOSTReplaces (isAppend false) or appends to (isAppend true) the URL or text list of a Loop Item action in a task. Read actionId with get_task_actions first. Professional or Enterprise plan.

Task analytics, 1 operation

query_task_analyticsQuery task analyticsPOSTReads per-task metrics over a date range in the last three months: rows collected, runs, success rate, run time and resources used (CAPTCHAs, proxy IPs, cost), by day, week or month.

Task group, 1 operation

list_task_groupsList task groupsGETLists all task groups (folders) of the account: taskGroupId and taskGroupName. Pass a taskGroupId to list_tasks_in_group. Standard, Professional or Enterprise plan.

Template, 1 operation

search_templatesSearch templatesGETStep 1 of a new scrape: find a ready-made Octoparse scraping template by keyword, id or slug. Pass the chosen templateName and inputSchema field keys to run_template.

Use Octoparse from Claude Code, Cursor or Codex

Connect Octoparse in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Octoparse operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Octoparse

Create your workspace
Sign up free. Nothing to install.
Connect Octoparse once
Add your Octoparse connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.