> ## Documentation Index
> Fetch the complete documentation index at: https://composio-preview.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Olostep

> Olostep provides APIs for web scraping, crawling, search, structured answers, monitoring, and data extraction.

![Olostep logo](https://mintcdn.com/composio-preview/gqjBKmmJCINafaiK/images/toolkits/olostep.svg?fit=max\&auto=format\&n=gqjBKmmJCINafaiK\&q=85\&s=7fe71d18fbd8cabe0e2e26fb83a6fca8)

* **Category:** web scraping & browser
* **Auth:** API\_KEY
* **Composio-managed OAuth available?** N/A
* **Tools:** 16
* **Triggers:** 0
* **Slug:** `OLOSTEP`
* **Version:** 20261101\_00

## Tools

> [accordion] Answer Question
>
> **Slug:** `OLOSTEP_ANSWER_QUESTION`
>
> Synchronously research a bounded task by searching and browsing the web, then return the completed answer and sources when available. This usage-metered operation usually takes 3-30 seconds and has no documented fixed dashboard-key credit price; check the account balance before expensive use.
>
> #### Input Parameters
>
> **task** `string` *required*
>
> Question or bounded research task to answer by searching and browsing the web.
>
> **json\_format** `string`
>
> Desired structured output. Provide either an example JSON object with empty values or a textual description of the required data shape.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Create Monitor
>
> **Slug:** `OLOSTEP_CREATE_MONITOR`
>
> Create a scheduled page or web-query monitor for asynchronous provisioning, with optional alert channels and a separate webhook. The schedule defaults to every hour.
>
> #### Input Parameters
>
> **query** `string` *required*
>
> Natural-language description of what to monitor.
>
> **webhook** `object`
>
> Webhook delivery configuration, separate from notification channels.
>
> **metadata** `object`
>
> User-controlled metadata object attached to the monitor.
>
> **frequency** `string`
>
> Natural-language UTC schedule, up to 50 characters. Defaults to every hour; the minimum supported interval is every 10 minutes.
>
> **notification** `object`
>
> Events and email, Slack, or SMS channels used for notifications.
>
> **output\_schema** `object`
>
> JSON Schema object for structured values extracted by each monitor run.
>
> **source\_policy** `object`
>
> Optional URL and domain allow or deny lists for monitor sources.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Get Credit Balance
>
> **Slug:** `OLOSTEP_GET_CREDIT_BALANCE`
>
> Return the connected team's remaining Olostep credits, credit lots, active subscription, and whether usage is enabled without consuming web-processing credits.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Get Operation Result
>
> **Slug:** `OLOSTEP_GET_OPERATION_RESULT`
>
> Retrieve a stored scrape, search, map, answer, crawl, or batch by its operation ID without repeating or rebilling the original work. Crawl and batch results may still report in\_progress.
>
> #### Input Parameters
>
> **operation\_id** `string` *required*
>
> ID returned when the selected operation was created, such as scrape\_..., search\_..., map\_..., an answer ID, a crawl ID, or a batch ID.
>
> **operation\_type** `string ("scrape" | "search" | "map" | "answer" | "crawl" | "batch")` *required*
>
> Kind of stored Olostep operation whose result should be retrieved.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] List Batch Items
>
> **Slug:** `OLOSTEP_LIST_BATCH_ITEMS`
>
> Return one page of completed or failed items from an asynchronous batch, including retrieve IDs, URLs, metadata, and extracted links.
>
> #### Input Parameters
>
> **limit** `integer`
>
> Maximum batch items to return in this page. Olostep recommends 10-50; smaller or larger positive values are accepted by the documented API.
>
> **status** `string ("completed" | "failed")`
>
> Return only completed or failed batch items.
>
> **batch\_id** `string` *required*
>
> Batch ID returned by Start Batch.
>
> **next\_cursor** `string`
>
> Opaque continuation cursor from a previous response; omit on the first call.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] List Crawl Pages
>
> **Slug:** `OLOSTEP_LIST_CRAWL_PAGES`
>
> Return one page of URLs and retrieve IDs processed by an asynchronous crawl, with a continuation cursor for additional pages. Use Retrieve Content for page bodies.
>
> #### Input Parameters
>
> **limit** `integer`
>
> Maximum crawl-page records to return in this response (1-50), subject to the provider's 10 MB response cap.
>
> **crawl\_id** `string` *required*
>
> Crawl ID returned by Start Crawl.
>
> **next\_cursor** `string`
>
> Opaque continuation cursor from a previous response; omit on the first call.
>
> **search\_query** `string`
>
> Optional query used to order returned pages by relevance.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] List Monitor Events
>
> **Slug:** `OLOSTEP_LIST_MONITOR_EVENTS`
>
> Return one page of a monitor's snapshot events, including change summaries and short-lived snapshot links.
>
> #### Input Parameters
>
> **limit** `integer`
>
> Maximum monitor events to return in this page (1-100).
>
> **monitor\_id** `string` *required*
>
> Monitor ID whose snapshot events should be listed.
>
> **next\_cursor** `string`
>
> Opaque continuation cursor from a previous response; omit on the first call.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] List Monitors
>
> **Slug:** `OLOSTEP_LIST_MONITORS`
>
> List the account's page and web-query monitors, including their IDs, lifecycle status, schedules, tracked targets, and notification settings.
>
> #### Input Parameters
>
> **include\_deleted** `boolean`
>
> Whether to include deleted monitors; omit to exclude deleted records.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Map Website
>
> **Slug:** `OLOSTEP_MAP_WEBSITE`
>
> Synchronously discover and optionally rank URLs on a website, returning one response page and a continuation cursor. This usage-metered operation consumes Olostep credits.
>
> #### Input Parameters
>
> **url** `string` *required*
>
> Website URL whose links should be discovered.
>
> **top\_n** `integer`
>
> Maximum number of top-ranked links to return when search\_query is provided.
>
> **next\_cursor** `string`
>
> Continuation cursor returned by a previous call; omit on the first call.
>
> **exclude\_urls** `array`
>
> URL-path glob patterns to exclude; exclusions override inclusions.
>
> **include\_urls** `array`
>
> URL-path glob patterns to include, such as /blog/\*\*.
>
> **search\_query** `string`
>
> Optional query used to rank discovered links by relevance.
>
> **include\_subdomain** `boolean`
>
> Whether to include URLs on subdomains of the supplied website.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Retrieve Content
>
> **Slug:** `OLOSTEP_RETRIEVE_CONTENT`
>
> Retrieve stored page content by retrieve ID in selected live-supported formats, including hosted URLs for oversized output.
>
> #### Input Parameters
>
> **formats** `array`
>
> Stored formats to return. Supported values are markdown, html, and json. Omit to return all available stored formats, including any screenshot or network-call artifacts captured at scrape time.
>
> **retrieve\_id** `string` *required*
>
> Content ID returned by a scrape, batch item, or crawl page.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Scrape Page
>
> **Slug:** `OLOSTEP_SCRAPE_PAGE`
>
> Synchronously scrape one web page and return requested content, links, metadata, a retrieve ID, and billing usage. This usage-metered operation can consume credits even when cached content is returned. Fresh scrapes, browser actions, extraction, and screenshots may cost more.
>
> #### Input Parameters
>
> **url** `string` *required*
>
> Public URL of the single page to scrape.
>
> **actions** `array`
>
> Browser actions to perform in order before collecting content.
>
> **country** `string ("US" | "CA" | "IT" | "IN" | "GB" | "JP" | "MX" | "AU" | "ID" | "UA" | "RU" | "RANDOM")`
>
> Residential country from which to load the page, or RANDOM.
>
> **formats** `array`
>
> Content formats to return for this scrape.
>
> **max\_age** `integer`
>
> Maximum acceptable cached-result age in seconds. Zero forces a fresh scrape; 604800 permits a result up to seven days old. Cached results can still consume credits.
>
> **parser\_id** `string`
>
> Parser ID for structured JSON extraction; include json in formats when using it.
>
> **transformer** `string ("postlight" | "none")`
>
> HTML cleanup transformer to apply.
>
> **links\_on\_page** `object`
>
> Options for extracting and filtering links on the page.
>
> **extraction\_schema** `object`
>
> Desired object schema for LLM-based structured extraction.
>
> **remove\_css\_selectors** `string`
>
> CSS removal setting: use default, none, or a JSON-stringified array of CSS selectors.
>
> **screenshot\_full\_page** `boolean`
>
> When screenshot format is requested, capture the full page after scrolling to the bottom.
>
> **wait\_before\_scraping** `integer`
>
> Milliseconds to wait after loading before collecting content.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Search Web
>
> **Slug:** `OLOSTEP_SEARCH_WEB`
>
> Synchronously search the web for relevant deduplicated links. This usage-metered operation costs 5 base credits per search; supplying scrape\_formats also charges for scraping each returned page. The response reports credits\_consumed when available.
>
> #### Input Parameters
>
> **limit** `integer`
>
> Maximum number of deduplicated links to return, from 1 to 25.
>
> **query** `string` *required*
>
> Natural-language web search query.
>
> **fast\_mode** `boolean`
>
> Use the query verbatim for lower latency instead of the broader default search pass.
>
> **scrape\_formats** `array`
>
> Scrape every returned link in these formats. Omit this field to avoid per-page scrape charges.
>
> **scrape\_timeout** `integer`
>
> Wall-clock seconds allowed for optional per-link scraping. Unfinished link content may be null.
>
> **exclude\_domains** `array`
>
> Bare domains to exclude from results.
>
> **include\_domains** `array`
>
> Bare domains to restrict results to. Olostep strips URL schemes and trailing slashes.
>
> **remove\_css\_selectors** `string`
>
> For optional link scraping, use 'default', 'none', or a JSON-stringified array of CSS selectors to remove.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Set Monitor State
>
> **Slug:** `OLOSTEP_SET_MONITOR_STATE`
>
> Pause future scheduled runs of an active monitor or resume scheduled runs of a paused monitor. Returns the updated monitor.
>
> #### Input Parameters
>
> **state** `string ("pause" | "resume")` *required*
>
> Desired monitor schedule state: pause future runs or resume them.
>
> **monitor\_id** `string` *required*
>
> ID of the Olostep monitor to pause or resume.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Start Batch
>
> **Slug:** `OLOSTEP_START_BATCH`
>
> Start an asynchronous, usage-metered scrape of 1 to 10,000 URLs and return the batch ID and initial status. This action does not wait for completion; use Get Operation Result to monitor the batch and List Batch Items to consume results.
>
> #### Input Parameters
>
> **items** `array` *required*
>
> Pages to scrape asynchronously, from 1 to 10,000 items.
>
> **parser** `object`
>
> Parser configuration applied to every item in the batch.
>
> **country** `string`
>
> ISO 3166-1 alpha-2 country code used for batch execution.
>
> **webhook** `string`
>
> Public HTTP(S) URL that receives a POST when the batch completes.
>
> **metadata** `object`
>
> Batch-level metadata. Supports at most 50 keys; keys are at most 40 characters without square brackets, and values are stringified scalars of at most 500 characters.
>
> **links\_on\_page** `object`
>
> Link extraction filters applied to every item in the batch.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Start Crawl
>
> **Slug:** `OLOSTEP_START_CRAWL`
>
> Start an asynchronous, usage-metered website crawl and immediately return its crawl ID and initial status. This action does not poll: use Get Operation Result to monitor status and List Crawl Pages to consume output. Work may continue for 1-10 minutes and can consume credits for up to max\_pages pages.
>
> #### Input Parameters
>
> **top\_n** `integer`
>
> With search\_query, crawl only the top N relevant links found on each page.
>
> **parser** `string`
>
> Parser name placed in scrape\_options.parser for every page. Olostep automatically adds json output when a parser is supplied.
>
> **formats** `array`
>
> Per-page output formats placed in scrape\_options.formats. HTML is always included automatically; raw\_pdf is not supported for crawls.
>
> **timeout** `integer`
>
> Stop background work after this many seconds and retain completed pages. Olostep may take about 10 additional seconds to stop.
>
> **webhook** `string`
>
> Public HTTP(S) URL that Olostep should notify when the crawl completes.
>
> **max\_depth** `integer`
>
> Maximum link depth from start\_url; 0 processes only the starting page.
>
> **max\_pages** `integer` *required*
>
> Maximum pages to process, from 1 to 100. This required integration bound limits usage-metered work; choose the smallest value that satisfies the task.
>
> **start\_url** `string` *required*
>
> Public HTTP(S) website URL where the crawl begins.
>
> **exclude\_urls** `array`
>
> URL-path glob patterns to exclude. Exclusions override include\_urls.
>
> **include\_urls** `array`
>
> URL-path glob patterns eligible for crawling.
>
> **search\_query** `string`
>
> Query used to find and rank relevant links during the crawl.
>
> **include\_external** `boolean`
>
> Also crawl first-degree links to external sites, which may increase usage.
>
> **follow\_robots\_txt** `boolean`
>
> Respect robots.txt restrictions. Disable only when authorized to do so.
>
> **include\_subdomain** `boolean`
>
> Include pages on subdomains of start\_url.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Update Monitor
>
> **Slug:** `OLOSTEP_UPDATE_MONITOR`
>
> Update a monitor's schedule, metadata, notification configuration, or webhook. Pass webhook as null to explicitly remove webhook delivery.
>
> #### Input Parameters
>
> **webhook** `object`
>
> Replacement webhook delivery configuration. Pass null explicitly to remove the existing webhook; omit this field to leave it unchanged.
>
> **metadata** `object`
>
> Metadata keys to merge into the monitor. Use an empty string value to delete an existing key.
>
> **frequency** `string`
>
> New natural-language UTC schedule. Changing it recreates the monitor's underlying schedule.
>
> **monitor\_id** `string` *required*
>
> Unique ID of the monitor to update.
>
> **notification** `object`
>
> Replacement notification events and email, Slack, or SMS channels.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not
