Hi everyone! I’m building a LinkedIn scraper Actor on Apify and have identified most of the endpoints I need. The remaining piece is retrieving a profile’s connection count through LinkedIn Classic. Has anyone worked on this or found an endpoint or reliable approach? I’d appreciate any insights and would be happy to discuss paid consulting for someone with relevant experience. Feel free to DM me if you can help. Thanks!

For each actor I have, I made a couple documentation pages. Generic one: https://bakosbence.com/web-data-collection/accela-permits/ Specialized one, how to guide for a given integration: https://bakosbence.com/web-data-collection/accela-permits/make/ And I Linked the generic one from py actor README(Documentation at the begining): https://apify.com/bakos_bence/accela-permits is that OK? I guess this would have a big seo value, linking back and forth.

Hi Apify team, Link previews (Discord, X, Slack) show no image for my recently published Actors. The og:image URL on the Actor page returns HTTP 400: "Categories must be comma separated list of valid categories" The URL is built with an empty categories param (…&categories=&…), because the legacy categories field is empty for new Actors now that categories are auto-assigned. Removing the empty param makes the image load fine. Older Actors with legacy categories are not affected. Affected, for example: https://apify.com/calm_builder/website-contact-scraper https://apify.com/calm_builder/instagram-profile-scraper Could you take a look? Thanks!

Hey! I see this on the monetization page(screenshot 1). I think I already did KYC, but whatever, I click the button anyways, but the KYC/KYB page sohws me a 400 error. (screenshot 2). I'm not even sure I'm supposed to get that alert box in the first place. https://cdn.discordapp.com/attachments/1550918503316594768/1550918504465694830/image.png?ex=6ab014ea&is=6aaec36a&hm=6f1b4cfa6ba6d693bef756bb99431b51e03a2c23deea77d1e21b8c9c611b09c2& https://cdn.discordapp.com/attachments/1550918503316594768/1550918507204706424/image.png?ex=6ab014ea&is=6aaec36a&hm=8802328184af191594f29de1f60bfcd4aadc60ba45c16b43737aa8ed868f1456&

when actor adds data to google sheets it deletes all available data https://cdn.discordapp.com/attachments/1550073085242376222/1550073085753958410/image.png?ex=6aad018e&is=6aabb00e&hm=fbad217122aa50acd8bc809c9a0e43b0f48569c3ba1df9045b425271a4bffb68&

Actor: https://apify.com/solutionssmart/actorwebapp What it does: turns a compatible Actor into a branded web app (form from input schema → run target Actor → show Dataset) via a shareable Standby URL, without exposing APIFY_TOKEN to the browser. Issue: - The real product path is Standby URL. - Console “Start” is only a health check and exits quickly (by design in the README). - In the Store this shows up as aborted / 0% runs succeeded, which looks broken to buyers even when Standby works. Questions: 1) For Standby-first Actors, what’s the recommended way to make “Start” / run history not look failed or aborted? 2) Should health-check Start exit SUCCEEDED with a clear message, stay RUNNING only under Standby, or is there a better pattern Apify wants for Standby apps? 3) Any Store/metrics tip so success rate reflects Standby usage rather than the health-check Start? Happy to share a run ID / Standby config (no tokens) if useful. Also curious for feedback on the whitespace idea itself: “Actor → instant branded frontend for customers/teammates.” If anyone from Apify has thoughts on Standby UX for this pattern, I’d love them.

I sent multiple claims to https://apify.com/ideas via the ideas@apify.com email in the past weeks, but my claims were not yet materialized. All the ideas on that page are open to develope, so I'm not even sure if I was the first one claiming given idea pages. The idea of the *idea*s page on apify is a very good idea in my opinion, so I hope the idea will materialize soon.

message has been deleted

2026-09-02T13:54:29.613Z INFO Data processed... 2026-09-02T13:54:29.613Z INFO 2026-09-02T13:54:29.614Z PHASE - UPLOADING TO SPREADSHEET 2026-09-02T13:54:29.614Z 2026-09-02T13:54:29.614Z Total rows: 3, total columns: 13 total cells: 39 2026-09-02T13:54:29.615Z Inserting new cells 2026-09-02T13:54:29.763Z ERROR Request Inserting new rows failed with error Requested writing within range [Sheet1!A40:J1000], but tried writing to column [K] 2026-09-02T13:54:29.764Z ERROR Requested writing within range [Sheet1!A40:J1000], but tried writing to column [K] 2026-09-02T13:54:29.764Z Requested writing within range [Sheet1!A40:J1000], but tried writing to column [K] 2026-09-02T13:54:29.765Z at Gaxios._request (/usr/src/app/node_modules/google-auth-library/node_modules/gaxios/build/src/gaxios.js:142:23) 2026-09-02T13:54:29.765Z at process.processTicksAndRejections (node:internal/process/task_queues:95:5) 2026-09-02T13:54:29.765Z at async OAuth2Client.requestAsync (/usr/src/app/node_modules/google-auth-library/build/src/auth/oauth2client.js:429:18) 2026-09-02T13:54:29.766Z at async exports.retryingRequest (/usr/src/app/src/utils.js:94:24) 2026-09-02T13:54:29.779Z at async module.exports (/usr/src/app/src/upload.js:20:9) 2026-09-02T13:54:29.780Z at async /usr/src/app/src/main.js:212:9 2026-09-02T13:54:29.780Z at async /usr/src/app/node_modules/apify/actor.js:167:23 https://cdn.discordapp.com/attachments/1544709416199258213/1544709416497188986/image.png?ex=6a997e40&is=6a982cc0&hm=b9cd6b83cfe839b1341b38349b149db3d62436ce3bf70836ff1d9f32706042c0& https://cdn.discordapp.com/attachments/1544709416199258213/1544709416983462020/image.png?ex=6a997e40&is=6a982cc0&hm=329c1c7f4edc299be04acec8a47a9e20b463ad2025e3fc1ab4985c030274224e&

If I'm correct, apify regularly does a smoke test on my public actors to ensure quality. I thought I will get a notification when this fails, but no. Can I set up notifiaction somehow? Is there a page dedicated for the purpose of checking which actors are failing?

message has been deleted

i need to transform fetched data (keep only some columns) , which parameters should i change. I send data to google sheet with integration. my data is like

[
  {
    "facebookUrl": "https://www.facebook.com/share/g/1DVsmsewEY/",
    "text": "",
    "user": {
      "id": "100011462833209",
      "name": "Xaliq Qarayev",
      "profilePic": "https://scontent-phl2-1.xx.fbcdn.net/v/t39.30808-1/786036778_2976986569360075_549C"
    },
    "likesCount": 0,
    "commentsCount": 0
  },
}]```i need to transform fetched data (keep only some columns)  , which parameters should i change. I send data to google sheet with integration. my data is like ```[
  {
    "facebookUrl": "https://www.facebook.com/share/g/1DVsmsewEY/",
    "text": "",
    "user": {
      "id": "100011462833209",
      "name": "Xaliq Qarayev",
      "profilePic": "https://scontent-phl2-1.xx.fbcdn.net/v/t39.30808-1/786036778_2976986569360075_549C"
    },
    "likesCount": 0,
    "commentsCount": 0
  },
}]

https://cdn.discordapp.com/attachments/1544044994891878530/1544044995630338138/apify_transform.PNG?ex=6a971375&is=6a95c1f5&hm=ad08bbe5539a36471d32803ffe52563dd548c66ec47b14f30510858463ceb36f&

message has been deleted

Opening the Output tab of a finished run shows an error instead of the data: Cannot load data from dataset. Reason: Expected property

clean

to be of type

boolean

but received type

string

in object The request the console makes is GET /v2/datasets//items?limit=50&offset=0&clean=1 - and the API is fine with that. I called it directly three ways: clean=1, clean=true, and with the parameter omitted. All three return 200 with the items. So this looks like client-side validation in the Console rejecting its own query string (clean=1 arrives as the string "1") before it renders, rather than anything wrong with the dataset or the API. Details: it happens on both a named dataset view and the "All fields" tab; reproduced in two different browsers on the same run; my .actor/dataset_schema.json never sets clean, the views only declare transformation.fields and display. Export and the API both work, so the data is fine and this is display only. Actor northbell/linkedin-jobs-salary-scraper, run iYekNlIEW4m9uvf5y. Flagging it because the Output tab is the first thing a customer sees after running an Actor, so right now they get an error where the results should be.

Hi @Apify Team / Support, Our paid account (username: bislokacuan) was abruptly disabled by the automated system today. * We renewed our subscription for $421.20 USD just 8 days ago (Aug 10). * No warning, specifics, or explanation regarding any alleged policy violation were provided. * All 3 requested verification proofs (LinkedIn + selfie + explanation) have already been submitted via email. The auto-responder mentioned a review timeframe of up to a week or two. A multi-week delay for an actively paid account completely halts our business workflow and is causing us major operational losses. Could a human staff member please escalate this case and help unblock our account as soon as possible? Thank you. https://cdn.discordapp.com/attachments/1539200599256469514/1539200599567110225/scrnli_3p3l8DUG3G1v6X.png?ex=6a8573c3&is=6a842243&hm=dde0a7748249d5d3bbbef53bdcf777e45a0f6dd534c1d4efd9950e20dd3750ac&

message has been deleted

Hey, I have a question about using Apify for long-term lead generation. I'm planning to use Apify to fetch around 1,000 leads per day based on specific locations and job positions. I expect to run this process continuously for about 4–5 months to build and enrich my database. My main concern is duplicate profiles across different runs. For example, let's say I've already collected 10,000 profiles. The next day, I request another 1,000 profiles matching the same criteria. How does Apify ensure that those 1,000 results haven't already been returned in previous runs? If 900 of those profiles are duplicates and only 100 are new, I would effectively be paying for 1,000 results while only adding 100 new leads to my database. Is there a built-in mechanism to prevent previously retrieved profiles from being fetched again across multiple runs? Ideally, I would like to maintain a list of already retrieved profile IDs/URLs and have Apify exclude them before scraping or charging for the results. Does Apify support something like this, or would the deduplication need to be handled on my side?

How can I get a review of a PR that's been sitting for a few weeks? https://github.com/apify/n8n-nodes-apify-template/pull/41

Has anyone got publish-task working with a multi-dataset actor (

storages.datasets

plural)? My actor uses two datasets in actor.json:

"storages": {"datasets": {"default": "./dataset_schemas/jobs_schema.json", "companies": "./dataset_schemas/companies_schema.json"}}

The Publication tab says "Your dataset schema isn't defined yet" and the Publish button stays greyed out. I've verified via the API that the build has the schema correctly stored -

actorDefinition.storages.datasets.default

has

fields

+

views.overview

, build log says "Found default dataset schema", dataset instances show

hasSchema: true

. Everything looks right server-side. Wondering if publish-task only supports

storages.dataset

(singular, one dataset) and not the plural multi-dataset form. Anyone hit this or found a workaround?

Hey @User — seeing similar issues to a few posts above, hoping you can help. My actor

factden/indeed-jobs-scraper

uses two datasets in actor.json: "storages": { "datasets": { "default": "./dataset_schemas/jobs_schema.json", "companies": "./dataset_schemas/companies_schema.json" } } The Publication tab → publish-task step says "Your dataset schema isn't defined yet" and the Publish button stays greyed out. I've confirmed via the API that the schema is correctly stored in the build — actorDefinition.storages.datasets.default has fields + views.overview, build log says "Found default dataset schema", dataset instances have hasSchema: true. Everything looks right except the Publication tab won't recognise it. I also found a Feb 2024 post in this channel where an Apify team member said "currently there is only one dataset under storages" for UI display. It looks like the publish-task feature was built on that old single-dataset infrastructure and hasn't been updated to support storages.datasets (plural/multi-dataset). Can you confirm this is the gap and whether there's a timeline for multi-dataset support in publish-task? Is storages.datasets (plural, multi-dataset form) supported by the publish-task feature? Or does it only work with storages.dataset (singular)?

message has been deleted

No new reviews added or anything, just suddenly changed the rating?? https://cdn.discordapp.com/attachments/1517944998375723088/1517944998669058178/image.png?ex=6a381ff1&is=6a36ce71&hm=d7fe28150029b89a2ea49048e037932ed6e839f4824ce562b50b4457b4932753&

Hi all, I'm currently dealing with a situation where a customer is requesting a refund. After using my actor, they made a technical mistake on their end which resulted in the same companies being searched multiple times, leading to a significant amount of duplicate charges. After reviewing their runs, I agreed to refund 50% of what they paid for those runs as a one-time exception. Before moving forward, I'd like to understand how Apify handles this on a platform level, so I have two questions: 1. What is the standard refund dispute resolution process on Apify? 2. If the customer contacts Apify support directly (via email or other channels), will Apify reach out to me for my input before making a decision? I'd really appreciate hearing from the Apify support team or anyone who has gone through a similar experience and can share how it was handled. Thanks in advance!

Hi Apify team/community, I’m trying to diagnose why my public Actors are hidden from default Apify Store search. Publisher: changeable_peddler The Actors are public/published, have complete Store metadata, public pages return 200, and a minimal test Actor has run successfully. But Store API default search returns non-zero totals/counts with empty or filtered items. Examples: GET /v2/store?search=store-indexing-control-actor&limit=50 returns total=1 count=1 items=[] GET /v2/store?search=changeable_peddler&limit=50 returns total=11 count=11 items=[] But if I add includeUnrunnableActors=true, the Actors appear, usually rank 1. This makes it look like Store safety/runnability filtering rather than indexing failure. We’ve already ruled out basic public/published metadata issues, missing runs, pricing, and Store page availability. Can someone from Apify confirm what backend/account/Actor eligibility flag usually causes this, and what action is needed to make these Actors appear in default Store search?

i am trying to connect to the openrouter apify actor through opencode "provider": { "apify-openrouter": { "npm": "@ai-sdk/openai-compatible", "name": "Apify OpenRouter Proxy", "options": { "baseURL": "https://openrouter.apify.actor/api/v1", "apiKey": "dummy-api-key", "headers": { "Authorization": "Bearer apify_api_2Zxx" } }, "models": { "anthropic/claude-sonnet-4.5": { "name": "Claude 4.5 Sonnet" } } } } but getting This request is not permitted in your current context. If you believe this is an error, please contact support.

Hey everyone 👋 I'm running into an intermittent 403 error on my actor when

Make.com

is connected to my Apify account. The error happens on

KeyValueStore.open()

at startup:

ApifyApiError: Insufficient permissions for the key-value store.
clientMethod: KeyValueStoreCollectionClient.getOrCreate
statusCode: 403
type: insufficient-permissions
httpMethod: post
path: /v2/key-value-stores

Context: - Running the actor on the Apify platform under an organization account - I am the owner of the org and the actor - The org API token has "Limit token permissions" turned OFF (full access) - When Make is disconnected → actor runs fine every time - When Make is reconnected → error reproduces intermittently (not every run, hard to reproduce) - Make is only making ~3 calls/day, so it's not spamming runs - The

KeyValueStore.open()

call uses a named store (

company-info-cache

) with get-or-create behavior - The store is properly dropped in the

finally

block at the end of every run Question: Is Make's OAuth connection somehow interfering with the token used during actor runs? Could reconnecting Make cause the platform to temporarily use Make's scoped token for actor runs triggered from the platform UI? Any ideas on what's causing this intermittent 403? Thanks!

message has been deleted

I would like to add a feature: I am using multiple datasets, but the UI experience is not very user-friendly—it doesn’t show how many records there are. It would be great if the

title

could be defined using variables, just like the

template

. Like this:

json
{
    “actorOutputSchemaVersion”: 1,
    “title”: “Output schema”,
    “properties”: {
        “products”: {
            “type”: “string”,
            “title”: “Products  ( {{storages.datasets.default.total}} items )”,
            “template”: “{{storages.datasets.default.apiUrl}}/items”
        },
        “categories”: {
            “type”: “string”,
            “title”: “Categories  ( {{storages.datasets.categories.total}} items )”,
            “template”: “{{storages.datasets.categories.apiUrl}}/items”
        }
    }
}

Output schema resources: https://docs.apify.com/platform/actors/development/actor-definition/output-schema

Can anyone advise on what the ideal settings are here? We want to publish an actor that user our internal API, so the Apify factor wont really be doing any heavy lifting. Any advise? https://cdn.discordapp.com/attachments/1509177972618428517/1509177973670940844/image.png?ex=6a183b04&is=6a16e984&hm=56fd2c83de48ed4caca9983156ede808c408b5d497eab9d07676a099480330c7&

Hello, I have been getting emails mentioning that our actors will be deprecated due to our actors not passing the automated testing. The issue is that our actors all require an API key generated for free (or paid) on our website in order to call the endpoints/actors. And some actors/endpoints are not available on our free plan but are available on our paid ones. How can I mitigate this? Rapidapi for example allows me to create tiers/plans that match our website so users can purchase from there + use on their platforms. But on apify I only see "pay per usage", "pay per event" and "pay per event + usage" What are my options? Can I supply a demo api key for the automated bot to use so it doesnt fail validation? and as for tiers/pay walled endpoints what are my options on that front? Here's a link to our pricing: https://vybe.fyi/api-pricing Thanks