Skip to main content
Glama
Remote Jobs MCP Server

Remote Jobs MCP Server

Official

Jobicy Remote Jobs MCP

Connect an AI assistant to Jobicy to find public remote jobs, read individual vacancies, discover companies and career roles, and resolve Jobicy filter names. The production MCP service is hosted at https://jobicy.com/mcp. This repository also contains a standalone Node.js MCP server for clients and hosting platforms that run a local process. It uses only public Jobicy API endpoints and does not contain the site's backend.

Connect your MCP client

Streamable HTTP URL: https://jobicy.com/mcp

In an MCP client that accepts a remote server URL, add a server named jobicy and paste that URL. If the client asks for a transport, select Streamable HTTP. No Jobicy account or access token is required for these public, read-only tools.

Clients that use a JSON MCP configuration can use this example:

{
  "mcpServers": {
    "jobicy": {
      "url": "https://jobicy.com/mcp"
    }
  }
}

Older clients that support only SSE can connect to https://jobicy.com/mcp/sse with transport type sse. The client handles the associated /messages route. The MCP discovery document is also available.

After connecting, ask your assistant to list available MCP tools. You should see search_jobs, get_job, and the other tools below.

Related MCP server: jobs-mcp

Run the repository server

Requires Node.js 20 or newer. Install dependencies with pnpm install --frozen-lockfile, then run node server.js. The process speaks MCP over standard input and output; it is intended to be launched by an MCP client or hosting platform. No Jobicy credentials are required. All job, company, role, and taxonomy data comes from the public Jobicy APIs.

A client that supports local MCP commands can use:

{
  "mcpServers": {
    "jobicy": {
      "command": "node",
      "args": ["/absolute/path/to/remote-jobs-mcp-server/server.js"]
    }
  }
}

For Glama's Dockerfile configuration, build the repository with pnpm install --frozen-lockfile and start the local process with mcp-proxy -- node server.js. Do not configure mcp-remote or the hosted /mcp URL as the container command: Glama's repository build runs the MCP process from this source. The live Jobicy endpoint remains available independently.

Find jobs

Use search_jobs for most searches. It returns compact vacancies without the full job description. For example, ask your assistant:

Find up to 10 remote backend engineering jobs open to candidates in the UK. Show the title, company, salary if available, and Jobicy link.

An equivalent tool call is:

{
  "name": "search_jobs",
  "arguments": {
    "query": "backend engineer",
    "geo": "uk",
    "limit": 10
  }
}

The response contains jobs, jobCount, nextCursor, hasMore, and appliedFilters. jobCount counts only jobs in that response, not all matching jobs. To request another page, pass nextCursor back as cursor with the same filters. Treat the cursor as an opaque value; changing filters invalidates it.

For one vacancy's complete description, call get_job with its numeric id. The geo filter describes candidate eligibility: a country search may also include jobs open to its broader region or to candidates anywhere. Salary and logo fields are included only when public data is available.

Available tools

Tool

When to use it

Inputs

search_jobs

Search compact public job results with cursor pagination.

Optional query, geo, industry, jobType, jobLevel, company, postedAfter, salaryMin, salaryMax, currency, limit, cursor.

get_job

Read one full public vacancy.

Required id.

get_similar_jobs

Find a short list of related vacancies.

Required jobId; optional limit.

search_companies

Find a specific public company or a small relevant set. This is a search, not a company directory.

At least one of query, industry, or geo; optional limit (up to 10).

get_company

Read one public company profile.

Required id (numeric ID or slug).

get_company_jobs

Find public jobs linked to one company.

Required companyId; optional limit, cursor.

get_role

Read a Jobicy career role guide.

Required role slug.

get_related_roles

Explore roles related to a career role.

Required role; optional limit.

resolve_taxonomy

Turn a location or industry name into a Jobicy filter slug.

Required type (location or industry) and query.

list_taxonomies

Browse available public filter slugs and labels.

Required type: locations, industries, job_types, or job_levels.

If a filter is obvious, call search_jobs directly. When unsure, use resolve_taxonomy: for example, United States resolves to usa, and Developer can resolve to engineering. If a match is uncertain, the tool returns candidates instead of guessing. Location labels in the new tools are shown without display emoji.

Legacy tools

Existing MCP clients can continue to use:

Tool

Inputs

Compatibility

get_jobs

count, geo, industry, tag; aliases location, category, search, keyword.

Keeps the original full jobs feed and JSON in content.text.

get_taxonomies

type: locations or industries.

Keeps the original JSON in content.text.

New tools provide both structuredContent and readable JSON in content. All tools are read-only. Responses do not provide global job or company totals or Jobicy internal statistics.

Resources

Clients that support MCP Resources can read individual public entities using jobicy://jobs/{id}, jobicy://companies/{id}, and jobicy://roles/{slug}. Public reference lists are available at jobicy://taxonomies/locations and jobicy://taxonomies/industries. Resources supplement the tools; they do not expose an all-jobs or all-companies feed.

Help and security

Open a GitHub issue for connection or documentation problems. Report security concerns privately as described in SECURITY.md.

License

This documentation is available under the MIT License.

Jobicy MCP server

Available Tools

12 tools
get_companyA
Read-only

Get public details of one company by stable ID or slug, including its description, website, categories and headquarters when available.

Inputs: id (required). Returns: company

get_company_jobsA
Read-only

Get compact public jobs for one published company. Cursor is opaque; pass it unchanged with the same companyId.

Inputs: companyId (required), limit, cursor. Returns: jobs, hasMore, jobCount, nextCursor, +1 more

get_jobA
Read-only

Fetch the full public description and details of one visible job by ID. Use after search_jobs when the complete vacancy is needed.

Inputs: id (required). Returns: job

get_jobsA
Read-only

Legacy-compatible public jobs feed. Keeps the original full job objects and JSON array in content.text. Use search_jobs for compact results and more filters. Geo means candidate eligibility: country, parent region, or globally available jobs may match. No authentication.

Inputs: geo, tag, count, search, +4 more. Returns: jobs

get_roleA
Read-only

Get a public Jobicy career role guide by its /careers/ slug, including existing overview, skills, responsibilities and levels.

Inputs: role (required). Returns: role

get_similar_jobsA
Read-only

Recommend a small set of public jobs related to one job. Deterministic score weights shared industry 3, type 2, level 2, location 1, title word 2. Scores are relevance weights, not probabilities.

Inputs: jobId (required), limit. Returns: jobs

get_taxonomiesA
Read-only

Legacy-compatible location or industry lookup. Returns the original JSON in content.text. Call when a filter slug is uncertain; obvious values can go directly to search_jobs.

Inputs: type (required). Returns: object

list_taxonomiesA
Read-only

List public filter slugs and labels for locations, industries, job types or job levels. No job or company counts are included.

Inputs: type (required). Returns: type, terms

resolve_taxonomyA
Read-only

Resolve an ordinary location or industry name to a real Jobicy slug. Use when unsure of a slug; low-confidence matches return candidates without selecting one.

Inputs: type (required), query (required). Returns: slug, type, label, query, +3 more

search_companiesA
Read-only

Find specific public companies by name, company industry, or headquarters text. A search criterion is required. Returns at most 10 companies, with no cursor or overall count; this is not a company directory export.

Inputs: geo, limit, query, industry. Returns: companies

search_jobsA
Read-only

Search public remote jobs with compact results, without full descriptions. Geo uses candidate eligibility, including eligible parent regions and global jobs. Call directly when filters are clear; use resolve_taxonomy for uncertain slugs. Cursor is opaque: pass nextCursor unchanged with identical filters. jobCount counts this page only.

Inputs: geo, limit, query, cursor, +8 more. Returns: jobs, hasMore, jobCount, nextCursor, +1 more

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 12 tools

    Disambiguation4/5

    Most tools target distinct resources and actions, and descriptions explicitly differentiate the legacy tools (get_jobs, get_taxonomies) from their modern counterparts (search_jobs, list_taxonomies). The taxonomy trio (get_taxonomies, list_taxonomies, resolve_taxonomy) and the two job-search tools still require reading the descriptions to distinguish, but the guidance is clear.

    Naming Consistency5/5

    All twelve tools follow a consistent verb_noun (or verb_noun_noun) snake_case pattern: get_job, search_jobs, list_taxonomies, resolve_taxonomy, get_company_jobs, etc. No mixed conventions or vague verbs.

    Tool Count5/5

    Twelve tools is well within the ideal 3-15 range and each maps to a distinct facet of the domain (jobs, companies, roles, taxonomies). No tool feels redundant enough to be padding, aside from the intentional legacy-compatible pair.

    Completeness4/5

    The read-only surface covers jobs (search, detail, similar), companies (search, detail, jobs), roles (guide, related), and taxonomy resolution/lookup, giving solid lifecycle coverage for a public jobs API. Minor gap: no bulk/aggregate or saved-search style operations, but these are reasonable omissions for a read-only public feed.

    Maintenance

    ActivityMaintained
    ResponsivenessLimited data

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      B
      maintenance
      Remote jobs MCP server — search 100,000+ remote jobs, post listings, find candidates, and check salary benchmarks from Claude, ChatGPT, Cursor, and any MCP client. Free, no API key required.
      21
      MIT
    • F
      license
      Not graded
      quality
      D
      maintenance
      Enables searching for job opportunities across multiple platforms like Upwork, RemoteOK, and GitHub while automatically generating tailored application proposals based on keyword scoring. It includes tools for scanning all sources simultaneously and managing a professional profile for quick reference during applications.
      1
      -
    • F
      license
      A
      quality
      D
      maintenance
      A Model Context Protocol server that enables AI agents to search, analyze, and apply for Upwork jobs autonomously. It facilitates specialized job scoring, proposal submission, and integrated client messaging through a combination of browser automation and API access.
      11
      14
      -
    • A
      license
      B
      quality
      C
      maintenance
      A fast, secure, and LLM-friendly Model Context Protocol (MCP) server that scrapes job listings from major platforms (LinkedIn, Indeed, Google) and converts them into structured Markdown format.
      1
      1
      MIT