How do I install the Linkedin Message Agent skill?

Run npx skillmds@latest add kevinbadi/linkedin-message-agent in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Linkedin Message Agent safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Linkedin Message Agent?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Linkedin Message Agent free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Linkedin Message Agent?

kevinbadi (@kevinbadi) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Linkedin Message Agent by kevinbadi

Name: Linkedin Message Agent
Author: kevinbadi

Linkedin Message Agent

Check LinkedIn inbox for new/unread messages and auto-respond using AI. Browser automation via CDP. Logs all conversations to database for tracking. Trigger when asked to check LinkedIn messages, respond to LinkedIn DMs, or monitor LinkedIn inbox.

kevinbadi 3423859 2 files · 22.1 KB Updated Sep 17, 2026

File contents

LinkedIn Message Check & Respond Agent

Monitors LinkedIn inbox for unread messages, reads conversation context, generates contextual AI responses, and sends replies. All via CDP browser automation.

Prerequisites

Chrome with CDP running (auto-starts via Clawdbot)
Logged into LinkedIn (cookies at cookies/linkedin-zeusbycreatoros.json)
ws and pg npm packages
DATABASE_URL env var for logging

Quick Usage

Check and respond to all unread messages

node skills/linkedin-message-agent/scripts/linkedin-message-agent.js

Check only (don't send responses)

node skills/linkedin-message-agent/scripts/linkedin-message-agent.js --dry-run

Limit number of conversations to process

node skills/linkedin-message-agent/scripts/linkedin-message-agent.js --limit 5

Workflow

Navigate to LinkedIn Messaging inbox
Scan for unread conversations (indicated by bold/unread indicators)
For each unread conversation:
- Click into the conversation
- Read the last few messages for context
- Determine if a response is needed (skip automated/spam)
- Generate a contextual response as Kev's Assistant (CreatorOS AI employee)
- Type and send the response
- Log the conversation to the database
Output summary of all processed conversations

Response Persona

Kev's Assistant responds as a professional AI employee at CreatorOS:

Friendly, helpful, concise
Interested in AI, automation, startups
Can discuss CreatorOS services if relevant
Avoids overselling, stays conversational
If someone asks to meet/call, suggests they connect with Kevin

CLI Options

Flag	Default	Description
`--dry-run`	false	Read messages only, don't respond
`--limit`	10	Max conversations to process
`--cdp-port`	18800	Chrome CDP port

Database

Logs to social_messages table:

platform: 'linkedin'
direction: 'inbound' or 'outbound'
sender_name, content, conversation context

Safety

Max 10 conversations per run (avoid spam detection)
5-10s delays between conversations
Never responds to obvious spam/automated messages
Dry-run mode to verify before enabling auto-respond

---name: linkedin-message-agentdescription: Check LinkedIn inbox for new/unread messages and auto-respond using AI. Browser automation via CDP. Logs all conversations to database for tracking. Trigger when asked to check LinkedIn messages, respond to LinkedIn DMs, or monitor LinkedIn inbox.---# LinkedIn Message Check & Respond AgentMonitors LinkedIn inbox for unread messages, reads conversation context, generates contextual AI responses, and sends replies. All via CDP browser automation.## Prerequisites- Chrome with CDP running (auto-starts via Clawdbot)- Logged into LinkedIn (cookies at `cookies/linkedin-zeusbycreatoros.json`)- `ws` and `pg` npm packages- `DATABASE_URL` env var for logging## Quick Usage### Check and respond to all unread messages```bashnode skills/linkedin-message-agent/scripts/linkedin-message-agent.js```### Check only (don't send responses)```bashnode skills/linkedin-message-agent/scripts/linkedin-message-agent.js --dry-run```### Limit number of conversations to process```bashnode skills/linkedin-message-agent/scripts/linkedin-message-agent.js --limit 5```## Workflow1. Navigate to LinkedIn Messaging inbox2. Scan for unread conversations (indicated by bold/unread indicators)3. For each unread conversation: - Click into the conversation - Read the last few messages for context - Determine if a response is needed (skip automated/spam) - Generate a contextual response as Kev's Assistant (CreatorOS AI employee) - Type and send the response - Log the conversation to the database4. Output summary of all processed conversations## Response PersonaKev's Assistant responds as a professional AI employee at CreatorOS:- Friendly, helpful, concise- Interested in AI, automation, startups- Can discuss CreatorOS services if relevant- Avoids overselling, stays conversational- If someone asks to meet/call, suggests they connect with Kevin## CLI Options| Flag | Default | Description ||------|---------|-------------|| `--dry-run` | false | Read messages only, don't respond || `--limit` | 10 | Max conversations to process || `--cdp-port` | 18800 | Chrome CDP port |## DatabaseLogs to `social_messages` table:- platform: 'linkedin'- direction: 'inbound' or 'outbound'- sender_name, content, conversation context## Safety- Max 10 conversations per run (avoid spam detection)- 5-10s delays between conversations- Never responds to obvious spam/automated messages- Dry-run mode to verify before enabling auto-respond

Frequently asked questions