Skip to main content
Glama
rcreech93

LinkedIn MCP

by rcreech93
OverviewSchemaRelated ServersScoreDiscussions
Remote

LinkedIn MCP

This is a simple implementation of an MCP LinkedIn server using a Selenium and Chromedriver based scraper in the background.

At the moment it only support profile scraping and profile search.

Pre-requisites

Please install uv.

Create a virtual environment and sync the libraries.

uv venv
uv sync

Then install Chromedriver: https://developer.chrome.com/docs/chromedriver/get-started

The whole parsing of the LinkedIn webpage is based on Chromedriver.

Related MCP server: LinkedIn MCP Server

Environment

Make sure you have all of these variables in your .env file:

LINKEDIN_USER_1=<user>
LINKEDIN_PASSWORD_1=<pass>

LINKEDIN_USER_2=<user>
LINKEDIN_PASSWORD_2=<pass>

COOKIE_DIR=/var/linkedin/cookies

MCP_HOST=0.0.0.0
MCP_PORT=8050
# Either sse or stdio or streamable-http
MCP_TRANSPORT=streamable-http
MCP_TIMEOUT=150

You can use multiple LinkedIn username and password combinations.

Run in dev mode with MCP Inspector

mcp dev ./src/linkedin_mcp/server/server.py

Run with stdio

python ./src/linkedin_mcp/client/client_stdio.py

Run the server

uv run ./src/linkedin_mcp/server/server.py

Note: If you're experiencing issues with code changes not being picked up, clear the Python cache first:

# Windows PowerShell
.\clear_cache.ps1

# Or manually:
Get-ChildItem -Path . -Filter __pycache__ -Recurse -Directory | Remove-Item -Recurse -Force

Then restart the server. Alternatively, use Python's -B flag to skip bytecode generation:

uv run python -B ./src/linkedin_mcp/server/server.py

Integration with Gemini CLI

This is a modified .settings.json of the Gemini CLI with which you can use for connecting and using the MCP:

{
  "general": {
    "preferredEditor": "vscode"
  },
  "security": {
    "auth": {
      "selectedType": "gemini-api-key"
    }
  },
  "ui": {
    "theme": "Default"
  },
  "mcpServers": {
    "linkedin-scraper": {
      "httpUrl": "http://0.0.0.0:8050/mcp/",
      "transport": "streamable-http"
    }
  }
}

Available Tools

2 tools
linkedin_profileA
Extract a LinkedIn profile using web scraping with Selenium.

Args:
    profile: The LinkedIn profile URL or username
    extract_educations: If True, extract educations
    extract_skills: If True, extract skills
    extract_interests: If True, extract the profile interests

Returns:
    The LinkedIn profile as a dictionary

Example output in case of success:
    {
        "given_name": "Gil",
        "surname": "Fernandes",
        "email": "gil-palma-fernandes@linkedin.com",
        "cv": "A Software Engineer and data scientist with over 25 years of professional experience in different companies. I spent a good part of my professional career working in projects related to the aviation industry and the rest as developer in Talend and Java based projects.

In 2019 I commenced my learning on Artificial Intelligence and Deep Learning and related technologies and participated in several Kaggle competitions. I dedicate my weekends to my pet projects for UK charities which promote well-being and meditation.", "summary": "", "industry_name": "Software Engineer at Onepoint Consulting", "geo_location": "London Area, United Kingdom", "linkedin_profile_url": "https://www.linkedin.com/in/gil-palma-fernandes", "experiences": [ { "institution_name": "Onepoint Consulting · Full-time", "linkedin_url": "https://www.linkedin.com/company/309728/", "website": "", "industry": "", "type": "", "headquarters": "", ...

Example output in case of error:
{
    "error": "Error extracting profile: Error extracting profile"
}
ParametersJSON Schema
NameRequiredDescriptionDefault
profileYes
extract_skillsNo
extract_interestsNo
extract_educationsNo

TDQS

A4/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The description reveals it uses web scraping with Selenium, which implies network/browser behavior, and includes an error example showing possible failure. However, it does not disclose potential issues like login requirements, rate limiting, or blocking, so transparency is moderate.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is well-structured with Args, Returns, and examples, making it scannable. The example output is lengthy but informative; while it could be trimmed, it mostly earns its place.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With no annotations and no output schema, the description compensates by detailing the return format with a comprehensive example and an error case. It covers all parameters but omits guidance on alternatives and edge-case behavior, making it largely complete but not exhaustive.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The description provides plain-language explanations for all four parameters, directly mapping to schema properties. The example output illustrates the effect of extraction flags, which is critical given the 0% schema description coverage.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description uses a specific verb 'Extract' and resource 'LinkedIn profile', clearly stating the tool's purpose. It distinguishes from sibling 'profile_search' by focusing on extracting a specific profile, not searching.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies usage when a specific LinkedIn profile URL or username is available, as shown by the 'profile' parameter. However, it does not explicitly contrast with 'profile_search' or state when not to use this tool, leaving guidance only implied.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

A4/5.0
Disambiguation5/5

The two tools have clearly distinct purposes: one extracts a full profile from a URL or username, while the other searches for profiles by name. There is no overlap or ambiguity in their intended use.

Naming Consistency3/5

The tool names do not follow a consistent verb_noun pattern. 'linkedin_profile' is a noun phrase, while 'profile_search' places the verb at the end. This is a minor inconsistency, but both names are descriptive and readable.

Tool Count3/5

With only 2 tools, the server feels slightly thin for a comprehensive LinkedIn integration. However, the two tools cover the core search-and-retrieve workflow, so the count is borderline appropriate for a focused profile-scraping server.

Completeness4/5

The tool set covers the main workflow of finding a profile via search and then retrieving its full details. Missing features like connections or posts are not critical for the apparent purpose of profile extraction, so the coverage is solid with only minor gaps.

Maintenance

ActivityInactive
ResponsivenessSyncing

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

Related MCP Servers

  • LinkedIn MCP Server

    markswendsen-code
    A
    license
    A
    quality
    D
    maintenance
    An MCP server for LinkedIn automation that enables users to search for jobs, retrieve profile details, manage connections, and read or send messages. It leverages Playwright and Browserbase to interact with LinkedIn through an existing authenticated browser session.
    6
    15
    1
    MIT

  • LinkedIn MCP Server

    selvin-paul-raj
    A
    license
    A
    quality
    D
    maintenance
    Fully featured MCP server that provides automation tools for LinkedIn, supporting browser-based scraping and API-based operations for content management, media uploads, and reactions.
    6
    3
    MIT

  • LinkedIn MCP

    Eldasoky1
    A
    license
    Not graded
    quality
    C
    maintenance
    MCP server for programmable LinkedIn automation via Playwright, offering 20 tools for profile management, messaging, feed interaction, and job searching through real browser automation.
    29
    MIT

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/rcreech93/linkedin_mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server