Skip to main content
Glama
ATS MCP Server

ATS MCP Server

by Pranali0809

ats-mcp-server

An MCP server over job boards. Ask it for a company's open roles and it answers from Greenhouse, Workday, or LinkedIn, normalized into one shape.

Greenhouse and Workday are public JSON feeds — no browser, no login, no API key. LinkedIn is different: it needs an authenticated browser session, so this server delegates to a running linkedin-mcp-server rather than reimplementing it.

Run it

pip install -e .

ats-mcp-server                      # stdio, for MCP clients that spawn a subprocess
ats-mcp-server --transport http     # streamable HTTP on 127.0.0.1:8100/mcp

Related MCP server: io.github.groundtruthtools/ats-jobs-mcp

Testing it

1. Without any setup — the public sources need nothing running:

python -m venv .venv && .venv/Scripts/python -m pip install -e .

.venv/Scripts/python -c "
import asyncio, json
from fastmcp import Client
from ats_mcp.server import mcp

async def main():
    async with Client(mcp) as c:
        r = json.loads((await c.call_tool('search_jobs',
            {'company': 'Stripe', 'ai_ml_only': True, 'limit': 5})).content[0].text)
        print(r['found_per_source'], '->', r['returned'], 'kept')
        for j in r['data']:
            print(' ', j['source'], j['title'])

asyncio.run(main())
"

2. With the MCP Inspector — a UI to click through every tool:

npx @modelcontextprotocol/inspector .venv/Scripts/ats-mcp-server

3. Including LinkedIn — start the delegated server first, in its own window:

uvx --with "fastmcp<4" "mcp-server-linkedin==4.23.1" \
    --transport streamable-http --host 127.0.0.1 --port 8000 --path /mcp

Then call source_status to confirm reachable: true, and pass sources: ["greenhouse", "workday", "linkedin"] to any search. LinkedIn is never searched unless you name it — see below.

4. From an MCP client (Claude Desktop, Claude Code) — point it at the stdio entry point:

{ "mcpServers": { "ats": { "command": "C:/path/to/.venv/Scripts/ats-mcp-server" } } }

Tools

Tool

What it does

find_board

Locate a company's board on Greenhouse or Workday. Cached for the process lifetime.

register_board

Pin a board explicitly when discovery can't guess it. Verified before it's accepted.

search_jobs

Search one company's roles, with optional AI/ML-only and junior filters.

search_many

Search several companies, merged into one list. Unresolvable names are reported, not fatal.

get_job

Full posting, including the description.

classify_title

Run the title heuristics without touching the network.

source_status

Which sources are configured, whether LinkedIn is reachable, quota left.

Why LinkedIn is opt-in

Greenhouse and Workday are searched by default. LinkedIn is not, and has to be named explicitly in sources.

Public feeds cost nothing to query. LinkedIn runs through a real logged-in account, where volume and machine-like regularity are what get accounts restricted. So it carries a throttle the public sources don't: a randomized 4-10s gap between calls and a self-imposed ceiling of 60 calls/day, both enforced in throttle.py. Hitting the ceiling returns a readable error rather than a platform-imposed restriction.

A source failing never fails the search. Ask for LinkedIn while its server is down and you still get the Greenhouse and Workday results, with the LinkedIn problem reported in errors.

How the sources differ

Greenhouse — one GET returns a company's entire board (Stripe's is ~630 postings). There's no search parameter, so the board is fetched once, cached for 15 minutes, and matched locally.

Workday — a POST with real server-side searchText, so a targeted query never pulls the whole board. Results are relevance-ranked by Workday, and how strictly it interprets a query varies by tenant: a small career site may return loosely-related postings. Titles are worth checking before trusting the set.

Board discovery

Greenhouse boards are addressed by a single token that's usually the company name (figma, stripe), so those resolve on the first guess.

Workday needs three coordinates — tenant, numbered host, and career-site name — and the site name is often company-specific (NvidiaExternalCareerSite). Discovery tries name-derived combinations first and gives up after a bounded number of probes rather than hammering a careers endpoint to establish that a company isn't on Workday. When it can't find one, read the coordinates off the careers URL and call register_board:

https://nvidia.wd5.myworkdayjobs.com/NvidiaExternalCareerSite/...
        └tenant┘ └───── host ─────┘ └────── site ──────┘

Filters

ai_ml_only keeps AI/ML/MLOps/GenAI/Data-Science titles and drops management and Data Engineer roles. Junior/intern/new-grad titles are excluded by default; include_junior keeps them. These are heuristics over titles, which is all a board search result usually gives you — classify_title shows how any title will be treated.

Notes

fastmcp is pinned >=4.0,<5 on purpose. It has removed decorator keyword arguments across majors before (4.0 dropped exclude_args), so an unbounded upper bound is a future crash on a fresh install.

Available Tools

7 tools
classify_titleClassify TitleA
Read-only

Judge one job title without touching the network.

Exposes the same heuristics the search filters use, so a caller can see how a title will be treated before trusting a filtered search.

Inputs: title (required). Returns: object

find_boardFind BoardA
Read-only

Find which job boards carry a company's postings.

Checks each source and reports every one that has the company, so a company on both Greenhouse and LinkedIn shows up as both. Results are remembered for the process lifetime.

Inputs: company (required), sources. Returns: object

get_jobGet JobA
Read-only

Full detail for one posting, including its description.

Inputs: job_id (required), source (required), company (required). Returns: object

register_boardRegister BoardA
Idempotent

Pin a board explicitly when discovery can't guess it.

Read the coordinates off the careers URL: https://boards.greenhouse.io/TOKEN, or https://TENANT.HOST/SITE/... for Workday. Verified before it's accepted, so a wrong value fails here rather than silently returning nothing later.

Inputs: source (required), company (required), host, site, +2 more. Returns: object

search_jobsSearch JobsA
Read-only

Search one company's open roles across job boards.

Public boards are searched concurrently. A source that has nothing for this company contributes nothing; a source that fails is reported in errors without losing the others.

Inputs: company (required), limit, query, sources, +2 more. Returns: object

search_manySearch ManyA
Read-only

Search several companies and return one merged list.

Companies run in sequence so a throttled source paces properly rather than firing every request at once. A company found on no board is listed in unresolved instead of failing the batch.

Inputs: companies (required), query, sources, ai_ml_only, +1 more. Returns: object

source_statusSource StatusA
Read-only

Which sources are configured, and what quota is left.

Public boards report as always available. LinkedIn reports whether the delegated server is reachable, plus its throttle state, so a caller can tell "not configured" apart from "quota spent".

Inputs: no named parameters. Returns: object

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.9/5.0

    Scored across 7 tools

    Disambiguation5/5

    Each tool has a clearly distinct role: discovery (find_board), manual pinning (register_board), single-company search (search_jobs), multi-company search (search_many), detail retrieval (get_job), local classification (classify_title), and source diagnostics (source_status). There is minimal overlap, and the descriptions clarify boundaries like one vs. many companies.

    Naming Consistency4/5

    Almost all tools use a consistent snake_case verb_noun pattern: find_board, register_board, search_jobs, search_many, get_job, classify_title. source_status breaks the pattern by starting with a noun, a minor inconsistency. Overall readable and predictable.

    Tool Count5/5

    Seven tools is well-scoped for an ATS/job-board search server; each tool covers a distinct operation without redundancy. The count supports core discovery, search, detail, classification, and status workflows.

    Completeness4/5

    The surface covers discovery, source registration, single/multi-company search, job detail, title classification, and source status, which are the main lifecycle operations for this domain. Minor gaps exist, such as no explicit board-removal or broader source-listing operation, but agents can work around these.

    Maintenance

    ActivityStale
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables users to discover applicant tracking system job boards from company domains, list open roles, detect hiring changes over time, and generate hiring summaries across multiple ATS platforms.
      MIT
    • A
      license
      A
      quality
      C
      maintenance
      Enables querying open job postings directly from company applicant-tracking systems (Greenhouse, Ashby, Lever), finding a company's job board, listing and comparing roles, and accessing salary data, all without scraping or API keys.
      3
      60 PyPI
      1
      MIT
    • A
      license
      A
      quality
      D
      maintenance
      Enables real-time job search across thousands of companies' open roles from Greenhouse, Lever, Ashby, and SmartRecruiters, with full-text filtering and company-specific queries, no API key required.
      2
      MIT
    • A
      license
      Not graded
      quality
      F
      maintenance
      Enables users to look up a company's open job postings from public Greenhouse, Lever, and Ashby boards, returning them in one normalized shape with filtering by keyword, location, remote status, and posting date. It also identifies which of those three ATSs hosts a given company's board and how many roles are open there.
      46 npm
      MIT