JobSpy MCP Server

<h1 align="center"> <a href="https://github.com/speedyapply/JobSpy"><img src="https://github.com/user-attachments/assets/dcb90340-9469-4808-bc51-868e59a000e0" alt="JobSpy" width="320"></a> </h1> <p align="center"> <a href="https://pypi.org/project/python-jobspy/"><img src="https://img.shields.io/pypi/v/python-jobspy" alt="PyPI version"></a> <a href="https://pypi.org/project/python-jobspy/"><img src="https://img.shields.io/pypi/pyversions/python-jobspy" alt="Python versions"></a> <a href="https://pepy.tech/projects/python-jobspy"><img src="https://static.pepy.tech/badge/python-jobspy/month" alt="Downloads per month"></a> <a href="https://github.com/speedyapply/JobSpy/actions/workflows/ci.yml"><img src="https://github.com/speedyapply/JobSpy/actions/workflows/ci.yml/badge.svg?branch=main" alt="CI"></a> <a href="https://github.com/speedyapply/JobSpy/blob/main/LICENSE"><img src="https://img.shields.io/pypi/l/python-jobspy" alt="License: MIT"></a> <a href="https://discord.gg/v7WdHnrNcN"><img src="https://img.shields.io/discord/1227411661221269565?label=discord&logo=discord&logoColor=white" alt="Discord"></a> </p>

JobSpy is a job scraping library that aggregates jobs from popular job boards with one tool:

• Scrapes job postings from LinkedIn, Indeed, Glassdoor, ZipRecruiter, Bayt, Naukri, BDJobs & HelloWork concurrently
• Aggregates the job postings in a dataframe
• Proxy support to bypass blocking

<p align="center"> <a href="https://github.com/speedyapply/JobSpy"><img src="https://github.com/cullenwatson/JobSpy/assets/78247585/ec7ef355-05f6-4fd3-8161-a817e31c5c57" alt="JobSpy demo"></a> </p>

Installation

pip install -U python-jobspy

Python version >= 3.10 required

Usage

from jobspy import scrape_jobs

jobs = scrape_jobs(
    site_name=["indeed", "linkedin", "zip_recruiter", "glassdoor"],  # also "bayt", "naukri", "bdjobs", "hellowork"
    search_term="software engineer",
    location="San Francisco, CA",
    results_wanted=20,
    hours_old=72,
    country_indeed="USA",
    # fetch_description=True,  # read each job's page on boards whose search results have no description (slower)
    # proxies=["user:pass@host:port", "localhost"],
)
print(f"Found {len(jobs)} jobs")
print(jobs.head())
jobs.to_csv("jobs.csv", index=False)

Output

SITE           TITLE                             COMPANY           LOCATION                JOB_TYPE  INTERVAL  MIN_AMOUNT  MAX_AMOUNT  JOB_URL                                            DESCRIPTION
indeed         Software Engineer                 AMERICAN SYSTEMS  Arlington, VA, USA      None      yearly    150000      200000      https://www.indeed.com/viewjob?jk=5e409e577046...  THIS POSITION COMES WITH A 10K SIGNING BONUS!...
indeed         Senior Software Engineer          TherapyNotes.com  Philadelphia, PA, USA   fulltime  yearly    110000      135000      https://www.indeed.com/viewjob?jk=da39574a40cb...  About Us TherapyNotes is the national leader i...
linkedin       Software Engineer - Early Career  Lockheed Martin   Sunnyvale, CA, USA      fulltime  yearly    None        None        https://www.linkedin.com/jobs/view/3693012711      Description:By bringing together people that u...
linkedin       Full-Stack Software Engineer      Rain              New York, NY, USA       fulltime  yearly    None        None        https://www.linkedin.com/jobs/view/3696158877      Rain’s mission is to create the fastest and ea...
zip_recruiter  Software Engineer - New Grad      ZipRecruiter      Santa Monica, CA, USA   fulltime  yearly    130000      150000      https://www.ziprecruiter.com/jobs/ziprecruiter...  We offer a hybrid work environment. Most US-ba...
zip_recruiter  Software Developer                TEKsystems        Phoenix, AZ, USA        fulltime  hourly    65          75          https://www.ziprecruiter.com/jobs/teksystems-0...  Top Skills' Details• 6 years of Java developme...

Command line

The package installs a jobspy command whose options mirror the scrape_jobs() parameters below; jobspy -h lists them, the output fields and exit codes.

uv tool install python-jobspy

jobspy "software engineer" --site indeed,linkedin --location Chicago -n 20
jobspy nurse --site indeed,linkedin --hours-old 24 -o jobs.jsonl
jobspy "data analyst" --remote --fields title,company,job_url

MCP server

mcp-jobspy is an MCP server that runs with uvx mcp-jobspy@latest. See mcp/README.md for setup, tools and proxies.

Standard config works for most clients:

{
  "mcpServers": {
    "jobspy": {
      "command": "uvx",
      "args": ["mcp-jobspy@latest"]
    }
  }
}

Parameters for scrape_jobs()

Optional
├── site_name (list|str): 
|    linkedin, zip_recruiter, indeed, glassdoor, google, bayt, bdjobs, naukri, hellowork
|    (default is all)
│
├── search_term (str)
|
├── google_search_term (str)
|     search term for google jobs. This is the only param for filtering google jobs.
│
├── location (str)
│
├── distance (int): 
|    in miles, default 50
│
├── job_type (str): 
|    fulltime, parttime, contract, temporary, internship, perdiem, other, volunteer
│
├── proxies (list|str): 
|    in format ['user:pass@host:port', 'localhost'], or a single proxy as a string
|    each job board scraper will round robin through the proxies
|    socks proxies on LinkedIn, Indeed and BDJobs need pip install "requests[socks]"
|
├── is_remote (bool)
│
├── results_wanted (int): 
|    number of job results to retrieve for each site specified in 'site_name'
│
├── easy_apply (bool): 
|    filters for jobs that are hosted on the job board site
|
├── user_agent (str): 
|    override the default user agent which may be outdated
│
├── description_format (str): 
|    markdown, html, plain (Format type of the job descriptions. Default is markdown.)
│
├── offset (int): 
|    skips that many results (e.g. 25 skips the first 25 and starts from the 26th)
│
├── hours_old (int): 
|    filters jobs by the number of hours since the job was posted 
|    (Glassdoor filters by whole days, so it rounds up to the next day.)
│
├── verbose (int) {0, 1, 2}: 
|    Controls the verbosity of the runtime printouts 
|    (0 prints only errors, 1 is errors+warnings, 2 is all logs. Default is 1.)
│
├── fetch_description (bool): 
|    for boards whose search results don't include the job description: fetches each job's
|    page for the description and other details (e.g. job type). Without it these are empty
|    for those boards. Adds one request per job, so use proxies for larger searches
│
├── linkedin_fetch_description (bool): 
|    deprecated, use fetch_description (still works; removed in 2.0)
│
├── linkedin_company_ids (list[int]): 
|    searches for linkedin jobs with specific company ids
|
├── country_indeed (str): 
|    filters the country on Indeed & Glassdoor (default is usa; see below for the names)
|
├── enforce_annual_salary (bool): 
|    converts wages to annual salary
|
├── ca_cert (str)
|    path to CA Certificate file for proxies

Supported Countries for Job Searching

LinkedIn

LinkedIn searches globally & uses only the location parameter.

ZipRecruiter

ZipRecruiter searches for jobs in US/Canada & uses only the location parameter.

Indeed / Glassdoor

Indeed & Glassdoor support most countries, chosen with the country_indeed parameter (default usa). Additionally, use the location parameter to narrow down the location, e.g. city & state if necessary.

You can specify the following countries when searching on Indeed (use the exact name, * indicates support for Glassdoor):

Argentina*	Australia*	Austria*	Bahrain
Belgium*	Brazil*	Canada*	Chile
China	Colombia	Costa Rica	Czech Republic
Denmark	Ecuador	Egypt	Finland
France*	Germany*	Greece	Hong Kong*
Hungary	India*	Indonesia	Ireland*
Israel	Italy*	Japan	Kuwait
Luxembourg	Malaysia*	Mexico*	Morocco
Netherlands*	New Zealand*	Nigeria	Norway
Oman	Pakistan	Panama	Peru
Philippines	Poland	Portugal	Qatar
Romania	Saudi Arabia	Singapore*	South Africa
South Korea	Spain*	Sweden	Switzerland*
Taiwan	Thailand	Turkey	Ukraine
United Arab Emirates	UK*	USA*	Uruguay
Venezuela	Vietnam*

Bayt

Bayt searches all countries unless location names one it covers, e.g. "Dubai, UAE" or "Saudi Arabia".

BDJobs

BDJobs searches Bangladesh. location takes a division or district, e.g. "Dhaka" or "Chattogram Division".

Naukri

Naukri searches India. location takes a city, e.g. "Pune".

HelloWork

HelloWork searches France. location takes a French city, postcode or region, e.g. "Lyon". Most jobs are posted in French, so French search terms find more.

Notes

• Indeed is the best scraper currently with no rate limiting.
• Indeed's date_posted is the day a job was added to Indeed, the same date hours_old filters on.
• Most job board endpoints are capped at around 1000 jobs on a given search.
• LinkedIn rate limits bursts of requests with a 429, which clears within about a minute. Use proxies for larger searches.
• Glassdoor rate limits above about 2 requests a second.

Frequently Asked Questions

Q: How do I tell a blocked board from one with no jobs?
A: Failures and ignored options are logged as they happen, and the result's attrs carry the same messages so your code can read them:

jobs.attrs["errors"]    # {"linkedin": "response status code 429 (blocked: use a proxy)"}
jobs.attrs["warnings"]  # {"bdjobs": ["job_type internship isn't supported, ignoring it"]}

---

Q: Why is Indeed giving unrelated roles?
A: Indeed searches the description too.

• use - to remove words
• "" for exact match

Example of a good Indeed query

search_term='"engineering intern" software summer (java OR python OR c++) 2025 -tax -marketing'

This searches the description/title and must include software, summer, 2025, one of the languages, engineering intern exactly, no tax, no marketing.

---

Q: No results when using "google"?
A: Google Jobs is currently unavailable.

---

Q: Received a response code 429?
A: The board blocked you for sending too many requests. All of the job board sites are aggressive with blocking. We recommend:

• Wait some time between scrapes (site-dependent).
• Try using the proxies param to change your IP address.

---

JobPost Schema

JobPost
├── id
├── site
├── job_url
├── job_url_direct
├── title
├── company
├── location
├── date_posted
├── job_type: fulltime, parttime, internship, contract, temporary, ...
├── salary_source: direct_data, description (parsed from posting)
├── interval: yearly, monthly, weekly, daily, hourly
├── min_amount
├── max_amount
├── currency
├── is_remote
├── job_level
├── job_function
├── listing_type
├── emails
├── description
├── company_industry
├── company_url
├── company_logo
├── company_url_direct
├── company_addresses
├── company_num_employees
├── company_revenue
├── company_description
├── skills
├── experience_range
├── company_rating
├── company_reviews_count
├── vacancy_count
└── work_from_home_type

A column is empty when the job board doesn't provide it.