ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified

linkedin-management

Automates LinkedIn interactions like posting, editing, deleting, and reacting by utilizing browser-captured session cookies and network request structures.

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/bjr-developer/linkedin-auto-post-publish
Or

LinkedIn Management Skill

This skill allows agents to automate interactions with LinkedIn by intercepting internal API requests (Voyager and SDUI) and mimicking them using Python's requests library.

When to use this skill

Use this skill when you need to programmatically:

  • Create text or image posts on LinkedIn.
  • Edit existing posts.
  • Delete posts.
  • React to posts (Like, Celebrate, Support, Love, Insightful, Funny).
  • Comment on posts.

How it works

LinkedIn uses a combination of GraphQL (Voyager) and Server-Driven UI (SDUI) endpoints. Authentication is primarily handled via cookies (li_at, JSESSIONID) and headers.

IMPORTANT: This skill uses placeholders for all session data. You must provide your own linkedin_cookies.json and session-specific headers.

Gathering Cookies & Session Data

To use this skill, you must manually gather:

  1. Cookies: Use a Chrome extension (like "Cookie-Editor") or the DevTools Application tab to export li_at and JSESSIONID. Save them in linkedin_cookies.json using the provided template.
  2. Session Headers: Use the included advanced-network-capture extension to find your x-li-page-instance when performing an action on LinkedIn. Update this value in li_interact.py.

Scripts Included

Located in the scripts/ directory:

1. capture_session.py

Uses Playwright to open a LinkedIn session, allowing the user to log in and perform actions manually. It captures the cookies and network logs. Usage:

python scripts/capture_session.py

2. li_interact.py

A comprehensive utility class LinkedInBot that handles automated actions. Supported actions:

  • create_post(text, media_urn=None)
  • edit_post(share_urn, activity_id, text)
  • delete_post(activity_id)
  • react_to_post(activity_id, reaction_type)
  • create_comment(activity_id, text)
  • upload_image(image_path)

Usage via Python:

from scripts.li_interact import LinkedInBot

bot = LinkedInBot(cookies_file="linkedin_cookies.json")
bot.create_post("Hello from automated LinkedIn bot!")

Setup Requirements

  • Python >= 3.10
  • Packages: requests, playwright
  • Cookies: li_at and JSESSIONID are mandatory.

[!WARNING] Automation on LinkedIn is strictly regulated. Use this skill sparingly and avoid high-frequency automated actions to prevent account suspension.

Metadata

Stars3376
Views0
Updated2026-03-24
View Author Profile
AI Skill Finder

Not sure this is the right skill?

Describe what you want to build — we'll match you to the best skill from 16,000+ options.

Find the right skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-bjr-developer-linkedin-auto-post-publish": {
      "enabled": true,
      "auto_update": true
    }
  }
}
Safety NoteClawKit audits metadata but not runtime behavior. Use with caution.