Linktree Profile Scraper

Returns: one row per public Linktree profile with username, name, bio, verified status, social links, destination links, link count, emails, phones, and external domains. Input: {"usernames":["MrBeast"],"maxItems":1} Best for: creator research, social account discovery, and monitoring links on public profile pages.

{"id":"e59389a3-1311-4f1c-befb-69eb623a5b28","sourceUrl":"https://linktr.ee/MrBeast","username":"MrBeast","name":"MrBeast","verified":true,"linkCount":15}

What does Linktree Profile Scraper do?

Linktree Profile Scraper reads public link-in-bio profile pages on linktr.ee and returns one structured row for each profile it finds. It takes usernames, full profile URLs, or both. The Actor extracts information embedded in the public page, including profile identity, a description, verification status, the avatar URL, social links, and the page's destination links. It also computes a link count and derives email addresses and social handles when those values are visible in the public page data. No login is required.

A Linktree profile can contain many links, social icons, and link groups. This Actor keeps those together in one profile row so a run with ten profiles can produce up to ten rows. The links array includes child links from groups. The socialLinks array is separate from ordinary destination links because Linktree represents it separately in the public page data. Empty arrays mean the public profile did not expose values in that category at the time of collection.

The Linktree scraper visits each profile page sequentially. It waits between requests and retries temporary blocking responses with a new proxy session. A missing username is logged and skipped. Each result includes the URL of the page used to create it and a stable account identifier. The output is suitable for repeat collection because those two fields let you compare results between runs.

This Actor is an independent tool and is not affiliated with Linktree. Public pages can change, disappear, or restrict visible links. The Actor reports what it can find in the page response during the run; it does not obtain private profile information or data behind a login.

What data can you get?

The dataset has one row for each public profile successfully read. Fields can be null or empty when a profile does not expose the corresponding information. In particular, a profile may show no ordinary links or no social links. The presence of a profile row does not imply every optional field is filled.

Field	Type	Meaning
id	string	Stable Linktree account identifier used to deduplicate a run.
sourceUrl	string	Profile page URL requested for the row.
username	string	Username shown in the public page data.
name	string	Public display name or page title.
bio	string or null	Public profile description, when present.
verified	boolean or null	Public verification flag supplied by the page.
avatarUrl	string or null	Public avatar image URL, when supplied.
createdAt	string or null	Account creation time normalized to an ISO date.
socialLinks	array	Public social icons as objects with type and url.
links	array	Visible destination links, including links inside groups, with title, URL, type, and thumbnail.
linkCount	integer	Count of entries in the flattened links array.
emails	array	Addresses found in the bio or mailto: destination links.
phones	array	Phone numbers from tel: links, WhatsApp links, and international numbers written in the bio.
externalDomains	array	Distinct non-social website domains the profile links to, useful for finding a creator's own sites and stores.
socialHandles	object	Platform-to-handle map derived from public social and destination URLs.
category	string or null	Optional AI classification: creator, musician, business, brand, or other.

A link URL is recorded as it appears in the profile data. The Actor does not visit each destination website, verify that a destination still works, or replace tracking and short URLs with their final destinations. This distinction matters when a profile uses a redirect service. The thumbnail value may be null even when a destination link exists. The socialHandles map covers Instagram, TikTok, Facebook, X, YouTube, LinkedIn, Threads, Twitch, Snapchat, Pinterest, Telegram, GitHub, Spotify, Discord and WhatsApp, and is a convenience derived from URL paths, so check the original socialLinks and links arrays when exact account identity matters.

The emails field is deliberately narrow. It collects address strings from the public bio and explicit email links. It does not guess an owner's email address or query an account directory. Many public profiles have no email address, and an empty array is a normal result.

How to use

Open the Actor's Input tab and enter one or more Linktree usernames, such as MrBeast. You may paste full https://linktr.ee/... profile URLs into the usernames field, or add them in the Start URLs field. Set Maximum profiles to the number of rows you want. Run the Actor, then open the Dataset tab to inspect or export the profile rows.

The input list is treated as a set of profiles: repeated usernames with different capitalization are collected once in a run. A run stops when it has pushed maxItems successful profile rows or exhausted the supplied list. If a supplied username is not found, it does not consume an output row. To get a specific number of rows, provide enough existing public profiles and set maxItems to the desired count.

For repeated monitoring, run the same input periodically and compare rows by id. The account identifier is a stronger comparison key than the display name, which an account owner can edit. Compare linkCount for a quick indication that the page changed, then inspect the links array to see which URLs or titles changed. Review sourceUrl alongside a changed record to confirm which profile was fetched.

The default proxy configuration uses US residential Apify Proxy sessions. A custom proxy configuration can be supplied through the Input tab. Requests are limited to about one page per second, and temporary blocks are retried with backoff. If a page remains inaccessible after the retry budget, it is skipped and a warning appears in the run log. A dead URL or a nonexistent username will not create a placeholder dataset item.

Input

Field	Required	Default	How it is used
usernames	One of usernames or startUrls	None	List of usernames or full linktr.ee profile URLs.
startUrls	One of usernames or startUrls	None	List of profile URL objects, each with a url property.
maxItems	No	50	Maximum successful profile rows to push, from 1 to 10,000.
proxyConfiguration	No	US residential Apify Proxy	Proxy settings used for page requests.
enrich	No	true	Ask Jev to classify owners when a TypeSafe key is configured.
maxJevCalls	No	500	Upper bound on Jev requests in one run.

Here is an input for two profiles. The maxItems value means at most two profile rows are stored. You may omit proxyConfiguration to use the Actor's residential default. The startUrls alternative is useful when your workflow already stores full profile URLs.

{
  "usernames": ["MrBeast", "selenagomez"],
  "maxItems": 2,
  "enrich": true
}

Only the linktr.ee host is accepted for profile URLs. A malformed profile entry is ignored when other valid entries remain; a run with no valid targets fails with a clear input error. There is no search endpoint in this Actor. You supply the profile identifiers you want to collect.

Output

The following abbreviated record is based on a local run against the public MrBeast profile. The full row includes the link and social-link arrays; the short version keeps the example readable. Values on live public pages can change between runs.

{
  "id": "e59389a3-1311-4f1c-befb-69eb623a5b28",
  "sourceUrl": "https://linktr.ee/MrBeast",
  "username": "MrBeast",
  "name": "MrBeast",
  "verified": true,
  "createdAt": "2025-08-15T05:37:11.000Z",
  "linkCount": 15,
  "category": null
}

A full links entry has a title, a URL, a Linktree link type, and a thumbnail URL or null. A socialLinks entry has a social type and URL. The linkCount number is calculated from the flattened links array, so nested child links count as separate entries when they have URLs. Other arrays may legitimately be empty. Some profiles show an avatar or verification badge but expose no ordinary destination links.

The optional category value is produced by TypeSafe Jev, a third-party AI processor. When enrichment is enabled and the TypeSafe key is available at runtime, the Actor sends the public name, bio, and a bounded selection of link titles and URLs for classification. If the processor returns an error or is unavailable, the profile row still appears with category: null. Profile extraction itself does not depend on that classification.

For verification workflows, a hidden __verifySnapshots flag saves source snippets for the first ten emitted records to the run's key-value store under SNAPSHOTS. It is intended for checking where output values came from and is not needed for ordinary runs. Snapshot entries are keyed by the dataset record ID.

How much does it cost?

This Actor is charged per result row pushed to the dataset. A successful AI classification can trigger an additional enriched-result event for that same row. The Actor does not charge a result event for a skipped, missing, or inaccessible profile. Open the Actor's Pricing tab for current event pricing and platform usage details.

Use maxItems to bound the number of result rows. The Actor stops when that count is reached, even if additional usernames remain in the input. If you want to check a small sample before collecting a larger list, set a small maxItems value and review the fields in the returned rows.

Use with AI agents (MCP)

Apify's MCP server can expose this Actor as a tool to an AI agent. Configure the server with the Store path so the agent can discover its input description and output fields:

{
  "mcpServers": {
    "apify": {
      "url": "https://mcp.apify.com/?tools=axiomworks/linktree-profile-scraper"
    }
  }
}

Ask an agent to collect a known public profile and summarize its links, or to compare a list of public profiles. For example: “Get the public Linktree profile for MrBeast and list its social accounts,” or “Collect these three Linktree usernames and show each profile's linkCount.” The agent should provide usernames or URLs because the Actor does not discover profiles by keyword. Another useful prompt is: “Return the sourceUrl, name, and emails fields for these known profiles.” Empty email arrays should remain empty rather than being filled by inference.

The MCP tool's input schema places target fields first and describes how to use the limits. Agent callers can set maxItems to make a run predictable. The data remains a snapshot of the public pages at collection time, so the agent should include the source URL when presenting a result to a user.

Integrations and API

You can send the same input through the Apify API, then read the run's dataset items. The examples below show the basic call shape. Replace the token placeholder with your own Apify token in your local environment. Keep credentials out of shared scripts, logs, and published repositories.

Python example:

import os
from apify_client import ApifyClient

client = ApifyClient(os.environ["APIFY_TOKEN"])
run = client.actor("axiomworks/linktree-profile-scraper").call(
    run_input={"usernames": ["MrBeast"], "maxItems": 1}
)
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["sourceUrl"], item["name"], item["linkCount"])

JavaScript example:

import { ApifyClient } from 'apify-client';
const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('axiomworks/linktree-profile-scraper').call({
  usernames: ['MrBeast'], maxItems: 1,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
for (const item of items) process.stdout.write(`${item.sourceUrl} ${item.name}\n`);

cURL example:

curl -X POST "https://api.apify.com/v2/acts/axiomworks~linktree-profile-scraper/run-sync-get-dataset-items" \
  -H "Authorization: Bearer $APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"usernames":["MrBeast"],"maxItems":1}'

The resulting JSON rows can feed a spreadsheet, a monitoring workflow, or an internal data pipeline. Keep id and sourceUrl when storing them so later runs can be matched reliably. To compare links, preserve the links array rather than relying only on linkCount; different link sets may have the same count.

FAQ

Why did a username return no row? The public page may not exist, may be temporarily blocked, or may not contain usable public profile data. Review the run log for the username and request status. A missing page is skipped, and the Actor does not push an error-shaped record.

Will one profile produce one row per link? No. Each successful profile produces one row. All visible destination links are nested in that row's links array, and linkCount gives the array length.

Why is category null? Classification requires a configured TypeSafe key and an available Jev response. If enrichment is disabled or a Jev request fails, category stays null and the profile row is still returned.

Can emails or socialHandles be empty? Yes. emails depends on explicit public text or email links. socialHandles depends on recognizable social-platform URLs. Empty collections are normal and should not be interpreted as missing data errors.

Does the Actor open destination links? No. It reads the Linktree profile page. Destination URLs are copied from that page and are not independently checked or followed.

Can I pass the same username twice? Yes, but it is deduplicated within the run by case-insensitive username. The output also carries a stable account id for cross-run matching.

Is it legal to scrape Linktree?

Public profile pages can be accessed without logging in, but laws and site rules differ by location and use. Review your intended use and the applicable rules before processing or redistributing collected information. The Actor only reads public page responses and does not access private pages, sign in, or bypass a CAPTCHA. It records public URLs and text supplied by the profile owner at the time of collection.

If you use this data in outreach or an automated decision, verify that your workflow has an appropriate basis and respects the people involved. Public availability alone does not establish permission for every downstream use. This tool is not affiliated with Linktree, and its output should be attributed to the public source URL when presented elsewhere.

Feedback

If a public profile produces an unexpected result, include the input username, the run identifier, the affected output field, and what the public profile displayed at the time. That context helps distinguish a changed page layout from a profile that intentionally leaves a field empty. Please avoid sending passwords, private account details, or API credentials in a report.

Feature requests are most useful when they describe a concrete output need, such as a link type that is visible on a public page but missing from the links array. For repeatability, note whether the page has groups, social icons, or a profile lock. The Actor's source URLs and stable IDs make it easier to locate and compare affected records.