LinkedIn Account Warmup Agent for Openclaw

An automated LinkedIn warming agent designed to mimic human behavior and safely grow new accounts through daily lightweight interactions.

weimingboya
v0.1.0
Mar 2, 2026
0
0
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install linkedin-warmup

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install linkedin-warmup using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is LinkedIn Account Warmup Agent?

This skill automates the process of warming up new LinkedIn accounts to establish credibility and reduce the risk of platform bans. By leveraging AdsPower fingerprint browsers and the Openclaw Skills framework, it performs human-like actions such as browsing feeds, viewing profiles, and sending a limited number of connection requests.

The agent operates in a low-frequency, high-randomization mode, ensuring that account activity remains within safe thresholds while gradually expanding the professional network. It focuses on the My Network suggestions to ensure a steady, organic growth pattern that avoids the common pitfalls of aggressive automation.

LinkedIn Account Warmup Agent Use Cases

  • Safely warming up newly created LinkedIn profiles to prevent immediate flagging.
  • Automating daily connection requests from the More suggestions for you section with human-like delays.
  • Maintaining a consistent social presence by simulating active feed browsing and profile interactions.
  • Managing multiple LinkedIn identities via AdsPower integration to eliminate environment fingerprinting risks.

How LinkedIn Account Warmup Agent Works

  1. The agent initiates a specific AdsPower browser profile via the Local API, configuring a custom CDP port for remote control.
  2. Openclaw Skills takes over the browser session (linkedin profile) and navigates to the LinkedIn homepage.
  3. It executes a warm-up phase by randomly scrolling the feed and spending time on specific posts to simulate active reading.
  4. The agent moves to the My Network page and scans the More suggestions for you area for potential connections.
  5. For each candidate, it performs a deep-view: visiting the profile, scrolling for 10-25 seconds, and then returning to the list to click Connect.
  6. The lifecycle concludes once the daily connection target is reached or the maximum duration timer expires, followed by an automated browser shutdown.

LinkedIn Account Warmup Agent Setup

To use this skill, ensure AdsPower is running with the Local API enabled. Configure your execution payload with your specific browser and API details:

{
  "adsPower": {
    "baseUrl": "http://127.0.0.1:50325",
    "apiKey": "YOUR_API_KEY",
    "userId": "YOUR_PROFILE_ID",
    "cdpPort": 30001
  },
  "browserProfile": "linkedin",
  "durationMin": 10,
  "connectTarget": 3
}

It is highly recommended to schedule this as a daily cron job during business hours in your target time zone.

LinkedIn Account Warmup Agent Data Schema & Taxonomy

The skill generates a structured summary of the session activity, which can be delivered via notification hooks like DingTalk or Slack:

Parameter Type Description
duration_total Integer Total runtime in minutes
connections_sent Integer Number of successful connection requests
connection_list Array List of names and profile URLs processed
risk_detected Boolean Whether a Captcha or account restriction was encountered
session_log String Narrative of feed interactions and scrolling activity

LinkedIn Account Warmup Agent Advanced Features

  • Intelligent risk detection that halts all operations immediately upon encountering Captchas or weekly invitation limits.
  • Headless and non-headless mode support for debugging or background execution.
  • Randomized behavioral delays (20-60 seconds) between connection requests to bypass pattern-based detection.
  • Deep profile interaction mode that views the target's full profile before sending a request to maximize acceptance rates within the Openclaw Skills ecosystem.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*