This security release features a fix for a critical severity security vulnerability.
\n\n\n\n...
","protected":false}` | Source summary HTML in `excerpt.rendered`; `excerpt.protected` is false. | | `excerptText` | string | Can be empty | `A short summary` | Plain text excerpt. | | `categories` | integer array or null | Empty for standard pages; null for invalid source taxonomy | `[14, 15]` | Category IDs. | | `tags` | integer array or null | Empty for standard pages; null for invalid source taxonomy | `[]` | Tag IDs. | | `categoryDetails` | object array or null | Empty for no categories; null for invalid taxonomy | `[{"id":2,"name":"News","slug":"news","link":"https://example.org/news"}]` | Each object has integer `id` and nullable string `name`, `slug`, `link`. Names/links come from embedded terms. | | `tagDetails` | object array or null | Empty for no tags; null for invalid taxonomy | `[]` | Same object shape as categoryDetails, for post tags. | | `extractionStatus` | string | Never null | `complete` or `partial` | Row status. Missing term names make a row partial. It is separate from the run status. | | `observedAt` | string | Never null | `2026-10-04T16:00:00.000Z` | UTC time when the row was processed. | #### KVS output and summary `OUTPUT` has `schemaVersion` (integer, 1), `status` (string), `items` (array of the same Dataset records) and `summary` (the RUN_SUMMARY object). None of these keys is null. | RUN_SUMMARY field | Type / empty value | Meaning / example | | --- | --- | --- | | `schemaVersion` | integer | `1`, contract version. | | `status` | string | `complete`, `partial`, `not_found` or `error`. | | `errorCode` | string or null | Input error such as `INVALID_SITE_URLS`; source errors are in outcomes. | | `stopReason` | string or null | Run-wide stop, such as `RESULT_LIMIT`, `REQUEST_LIMIT`, `TIME_LIMIT`, `CANCELLED`, `CHARGE_LIMIT` or `TOTAL_BYTES_LIMIT`. | | `rows` | integer | Saved content row count; equals Dataset length and OUTPUT.items length. | | `requests` | integer | Attempted HTTP requests, including redirects and retries. | | `retries` | integer | Retried temporary failures. | | `responseBytes` | integer | Cumulative serialized JSON response bytes; transport also caps raw response bytes. | | `skippedProtected` | integer | Private, non-published or protected records returned by a source and skipped. | | `skippedOtherTypes` | integer | Custom content types skipped, such as podcast records in a posts endpoint. | | `duplicates` | integer | Repeated site/type/ID records not saved twice. | | `elapsedMs` | integer | Runtime processing duration in milliseconds. | | `outcomes` | array | One result for each chosen site/type. | Each outcome has `sourceUrl` and `contentType` strings, `status` string, `errorCode` string or null, `rows` and `pages` integers, plus `total` and `totalPages` integers or null. Totals are source headers and can include source records outside the supported types. Missing headers stay null. Example: posts with `status=partial`, `errorCode=PAGE_LIMIT`, `rows=2`, `pages=1`. ### Pricing Pay per event: - **Start:** $0.0045 per GB of Actor memory, with at least one start event. The default 512 MB has one event. - **Saved record:** $0.0019 for each unique post or page saved to the Dataset. - **Platform usage:** charged to you by Apify separately. This includes compute, storage and traffic. A valid empty run or search with no matches has the start charge and platform usage, with no saved-record charge. Failed, partial or cancelled runs still have the start charge and charges for records already saved. The Actor does not charge for a skipped type, duplicate, summary or source error. Dataset writes and charging are separate services; if a write or charge acknowledgement is lost, the run stops and does not retry that row. Inspect the Dataset and summary before repeating a failed run. | Saved records, default memory | Event charge | Extra | | ---: | ---: | --- | | 0 | $0.0045 | Apify platform usage | | 1 | $0.0064 | Apify platform usage | | 10 | $0.0235 | Apify platform usage | | 100 | $0.1945 | Apify platform usage | | 1,000 | $1.9045 | Apify platform usage | The default maximum event cost per run is $2. You can set a lower cost limit in run options, if it is at least the start fee. A charge limit stops new saves and gives `partial`. The event limit is not a fixed total invoice: platform usage is separate. Higher memory increases start events. The runtime caps remain in place. ### Use cases | Job | Input | Useful fields | Next step | | --- | --- | --- | --- | | Content migration | Choose posts and pages | content.rendered, title.rendered, slug, link | Export JSON for your migration tool. | | SEO content list | Add a site and small limit | id, titleText, link, date, categories | Export CSV for review. | | Text collection | Choose posts and optional search | contentText, excerptText, sourceUrl | Read Dataset records in your text workflow. | ### Schedules and repeat runs You can create a schedule in Apify Console using the same input. Each run is a new snapshot with normal charges. There is no stored change state, new-only mode or unchanged discount. Use sourceUrl + type + id and modified_gmt to compare exports in your own system. A missing record in a bounded sample does not prove that the source deleted it. ### API, MCP and integrations Use the [Apify API](https://docs.apify.com/api/v2) to POST the same input to `acts/cliqtomedia~wordpress-posts-pages-scraper/runs`. Add your API token in an Authorization header in your own application. Keep it out of shared URLs and screenshots. Set memory, timeout and maximum event charge in run options. After the run, read the default Dataset and KVS records. For an agent workflow, the [Apify MCP server](https://docs.apify.com/platform/integrations/mcp) can call the Actor with the same input. Pass `link`, `titleText` or `contentText` from each Dataset row to the next step. The Dataset can be exported as JSON or CSV. In a Make or Zapier workflow, start the Actor with siteUrls, wait for completion, check RUN_SUMMARY.status, then read the Dataset rows. Check `partial` before treating the result as a full export. ### Limits and partial results The result cap is 1,000 saved records per run. A live WordPress News export returned 1,000 supported posts at 512 MB. This proves that sample, not every WordPress site. Sources can change, block cloud IPs, disable REST API routes or return large content. Runtime caps: 100 source HTTP attempts including redirects and retries; about 5 minutes of source processing; 15 seconds per request or DNS lookup; 3 redirects per request; 8 MiB per HTTP response; 64 MiB cumulative serialized JSON responses; 1 MiB per HTML field. Requests are sequential with a delay. A large or slow response can end the run before its result cap. `complete` means the selected collections were exhausted, including a valid empty result. `partial` means a cap, cancellation, missing taxonomy or failed collection left part of the result incomplete. `not_found` means all selected REST routes returned 404. `error` means input, access, network or parsing failed without usable rows. Error and not_found runs are marked FAILED by Apify; partial runs can be SUCCEEDED but must still be checked in KVS. HTML is returned as data. It can contain source forms, public transient fields and scripts. Plain text skips scripts/styles but is not a visual copy of the page. Dates keep WordPress's meaning; the Actor does not guess a site's timezone. Source values can change between page requests. It does not provide a point-in-time database backup or resume a cancelled run. Read the [WordPress pagination guide](https://developer.wordpress.org/rest-api/using-the-rest-api/pagination/) for the source page and total-header rules. ### Troubleshooting | Symptom | Possible reason | Check and next action | | --- | --- | --- | | 0 rows, complete | A valid collection is empty or search has no matches | Check search and selected types. Remove search for a small new run. | | 0 rows, error | Non-WordPress HTML, denied REST API or malformed JSON | Check outcomes.errorCode. Use the WordPress installation root and verify its public REST API. | | not_found | Both supported REST route formats return 404 | Check the site root or subdirectory. This Actor cannot enable the API. | | Some rows, partial | A limit, missing terms or one failed site | Check row extractionStatus and outcomes. Increase page/result caps only within the allowed range. | | HTTP_429 or HTTP_5xx | Source rate limit or server problem | The Actor retries temporary failures up to twice. Increase requestDelayMs or try a small run later. Long Retry-After values stop with a clear code. | | UNSAFE_ADDRESS or CROSS_SITE_REDIRECT | Private address or another destination host | Use a public site root. Private addresses and unrelated redirect hosts are rejected. | | RESPONSE_TOO_LARGE or CONTENT_TOO_LARGE | The source payload exceeds a size limit | Lower perPage for response size. A single oversized field remains unsupported. | | PAGE_LIMIT or RESULT_LIMIT | More source records exist than requested | The result is a bounded sample. Increase the matching cap to read more. | | REPEATED_PAGE | The source returned the same records again | Check the source API pagination. The Actor stops rather than collecting duplicates forever. | ### FAQ #### Do I need a WordPress login? No. The source API must be public. The Actor does not accept passwords or cookies. #### Can it collect private posts or drafts? No. It asks for `publish` in the public view and skips protected or non-published records. #### Does it work on every WordPress site? No. It needs a reachable core REST API on that site. Some hosted sites, plugins and firewalls change or block the API. #### Is an empty run free? No. It has the start event and Apify platform usage, but no saved-record event. #### Can it collect all content without a limit? No. Each run has explicit result, page, time, request and size limits. Check the summary before using an export. #### Are category and tag names always available? No. IDs can be present while embedded names are missing. Those values stay null and extractionStatus is partial. Standard pages have empty taxonomy arrays. ### Related Actor [RSS & Atom Feed Reader / Monitor](https://apify.com/cliqtomedia/rss-atom-feed-reader-monitor) — read a feed and track feed item changes when a feed fits your source better. ### Support Open the Actor's **Issues** tab. Include the run ID, content type and safe error code. Do not send API tokens, passwords or private source content. # Actor input Schema ## `siteUrls` (type: `array`): Add 1–5 public WordPress site roots or subdirectories. No login, URL query or private network addresses. Duplicates are removed. ## `contentTypes` (type: `array`): Choose posts, pages, or both. The chosen order uses the shared result limit. ## `maxResults` (type: `integer`): Global limit across all sites and content types. A limit stop is partial. ## `maxPagesPerType` (type: `integer`): Maximum API pages for each site and content type. Start with 1 for a small test. ## `perPage` (type: `integer`): WordPress page size. Start with 20; larger responses can hit the response size limit. ## `search` (type: `string`): Optional source-side search, applied to all selected content types. Empty means no search filter. ## `requestDelayMs` (type: `integer`): Delay before each API request. Keep 1000 or increase it for smaller sites. ## Actor input object example ```json { "siteUrls": [ "https://wordpress.org/news" ], "contentTypes": [ "posts", "pages" ], "maxResults": 100, "maxPagesPerType": 5, "perPage": 20, "search": "", "requestDelayMs": 1000 } ``` # Actor output Schema ## `records` (type: `string`): No description ## `output` (type: `string`): No description ## `summary` (type: `string`): No description # API You can run this Actor programmatically using our API. Below are code examples in JavaScript, Python, and CLI, as well as the OpenAPI specification and MCP server setup. ## JavaScript example ```javascript import { ApifyClient } from 'apify-client'; // Initialize the ApifyClient with your Apify API token // Replace the '