MCP Servers · stickerdaniel

linkedin-mcp-server

An open-source MCP server that integrates LinkedIn with Claude and other AI agents, allowing them to access profiles, companies, jobs, and messaging through your own authenticated browser session. Runs locally and uses headless Chromium to scrape LinkedIn data without official API access.

Talk to DEV.co Discuss Implementation

Source: GitHub — github.com/stickerdaniel/linkedin-mcp-server

2.7k

GitHub stars

472

Forks

Python

Primary language

Apache-2.0

License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

Field	Value
Repository	stickerdaniel/linkedin-mcp-server
Owner	stickerdaniel
Primary language	Python
License	Apache-2.0 — OSI-approved
Stars	2.7k
Forks	472
Open issues	85
Latest release	v4.17.0 (2026-06-28)
Last updated	2026-07-08
Source	https://github.com/stickerdaniel/linkedin-mcp-server

What linkedin-mcp-server is

Python-based MCP server using Patchright (headless browser automation) to interact with LinkedIn's web interface. Supports 16+ tools for profile queries, job search, messaging, and company information. Deploys via uvx, Docker, or local development; requires active browser session and handles authentication via imported or newly logged-in session.

Quickstart

Get the linkedin-mcp-server source

Clone the repository and explore it locally.

terminalbash

1git clone https://github.com/stickerdaniel/linkedin-mcp-server.git2cd linkedin-mcp-server3# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

AI-powered recruitment workflows

Claude can search for candidates, fetch profiles with detailed work history/skills, and retrieve company employee lists to automate sourcing and qualification tasks.

Job market research and aggregation

Agents can search jobs by keyword and location, retrieve full job details, and cross-reference company information to support career guidance or market analysis.

Conversation context and networking automation

AI can read inbox messages, search conversations, and draft outreach, enabling semi-automated relationship management and sales engagement workflows.

Implementation considerations

Requires uv package manager and Chromium browser; uvx auto-downloads on first run. Patchright cache (~/.linkedin-mcp/patchright-browsers) will consume several hundred MB.
Authentication is session-based: either import from local browser or log in interactively via headless window. Sessions persist but may require re-auth if LinkedIn detects anomalies.
Early tool calls may fail with setup/auth errors until browser initialization completes; plan for graceful retry or user feedback in agent flows.
Tool output varies in completeness: some fields are optional, nested sections require explicit selection in requests (e.g., `experience`, `education`, `skills`).
Message sending requires explicit confirmation; other destructive tools (connect_with_person, send_message) have open issues indicating reliability concerns.

When to avoid it — and what to weigh

Requires official LinkedIn data guarantees — This is a web scraper, not the official LinkedIn API. Data format, availability, and completeness depend on LinkedIn's page structure and are subject to change without notice.
High-volume or production-critical scraping — Browser-based scraping is slower and more fragile than API calls. LinkedIn may block or rate-limit sessions; the project carries 85 open issues including multiple blocking bugs.
Strict compliance or enterprise environments — Using a personal browser session to scrape LinkedIn may violate ToS and data-handling policies. Review Unipile or official APIs for regulated use cases.
Hands-off deployment without monitoring — Sessions expire, page structures change, and authentication can fail silently. Requires active maintenance and operator involvement to debug and recover.

License & commercial use

Apache License 2.0 (Apache-2.0). Permissive OSI-approved license allowing commercial use, modification, and redistribution with attribution and no warranty. See LICENSE file in repo.

Apache-2.0 permits commercial deployment. However, the software scrapes LinkedIn's web interface, which likely violates LinkedIn's ToS. Legal review is strongly advised before commercial use; unauthorized scraping carries ToS breach and potential legal risk. Consider official LinkedIn API or Unipile (mentioned sponsor) for compliant commercial solutions.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

Signal	Assessment
Maintenance	Active
Documentation	Adequate
License clarity	Clear
Deployment complexity	Moderate
DEV.co fit	Good
Assessment confidence	High

Security considerations

Session-based authentication stores LinkedIn credentials/cookies locally (~/.linkedin-mcp/patchright-browsers). Headless browser runs with network access and arbitrary JavaScript execution. No encryption of stored profiles mentioned. Risks: credential exposure if host is compromised, browser cache inspection by other users, and potential for malicious upstream package updates (uses @latest uvx resolution). Review host access controls and consider isolated environments for sensitive data.

Alternatives to consider

Official LinkedIn API (if available)

Provides official, rate-limited, and ToS-compliant access. Requires LinkedIn approval; limited to specific use cases. Stable and audited but slower approval and higher cost.

Unipile (mentioned sponsor)

Hosted LinkedIn API wrapper handling auth, rate-limiting, and infrastructure. Managed alternative to local scraping; free trial available. Eliminates session/ToS/compliance burden at cost of external dependency.

RapidAPI LinkedIn scrapers (third-party)

Proxy-based or pre-scraped LinkedIn data services. Varies in reliability and legal standing; check ToS compliance and data freshness before commercial use.

Software development agency

Build on linkedin-mcp-server with DEV.co software developers

Install via uvx in minutes. Review the GitHub repo, GitHub issues, and legal implications before production use.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

mcp mcp-server-browserbase mysql_mcp_server

Related on DEV.co

Explore the category and the services that help you build with it.

MCP Servers AI development API development CRM development Talk to DEV.co

linkedin-mcp-server FAQ

Will this work with Claude Desktop?

Yes. Add the MCP server config (uvx command shown in README) to Claude Desktop's mcp_config.json, and tools become available in Claude. Requires local uv and Chromium.

What happens if LinkedIn changes its page layout?

Tools will fail or return empty/wrong data. Fixes are pushed as releases (frequent); always use @latest in uvx config to get updates automatically. Pinned versions will quietly break.

Can I use this without logging in?

No. Requires authenticated session. Log in via headless browser (--login), import from local browser (--import-from-browser), or stored persistent profile from prior login.

Is this legal for commercial use?

License (Apache-2.0) permits it. However, scraping LinkedIn's web interface violates LinkedIn's ToS. Commercial use carries ToS and potential legal risk; consult legal counsel and consider official APIs or Unipile.

Software developers & web developers for hire

DEV.co helps companies turn open-source tools like linkedin-mcp-server into production software. Our software development services cover the full lifecycle — architecture, web development, integration, and maintenance — delivered by software developers and web developers who ship. Engage our software development agency to implement or customize it for your mcp servers stack.

Ready to connect Claude to LinkedIn?

Install via uvx in minutes. Review the GitHub repo, GitHub issues, and legal implications before production use.

Talk to DEV.co Schedule a Consultation