Himalayas MCP server

What it does

This server connects AI assistants to Himalayas.app, a remote jobs marketplace. It lets you search job listings, post openings, find candidates, check salary data, manage applications, and update your career profile through conversation.

• Search remote job listings by role, salary range, location, and skills to find positions matching your criteria.
• Post job openings by describing the role in natural language, including salary, employment type, and optional screening questions.
• Track your job applications in a kanban-style tracker with statuses, notes, and excitement ratings.

Tools this server exposes

Tool	Does
search_jobs	Search the Himalayas job marketplace with filters for salary, country, experience level, employment type, and skills.
get_job_details	Retrieve the full posting details, description, company information, and application link for a specific job.
get_related_jobs	Find similar remote job listings based on a given job.
get_salary_data	Get salary benchmarks for a job title by seniority and country, with minimum, median, and maximum salaries in USD.
post_job	Create and submit a job posting with salary range, employment type, skills, and screening questions.
post_job_public	Post a job listing without requiring a Himalayas account.
search_candidates	Search remote candidate profiles by keyword, country, and skills to find talent.
get_company_info	Retrieve company profiles including tech stack, benefits, open positions, employee count, and social links.
save_job	Add a job to your application tracker.
update_application_status	Change the status of a tracked job application (saved, applied, interviewing, negotiation, hired, archived).
get_my_applications	Retrieve all tracked job applications from your kanban-style tracker.
update_profile	Modify your Himalayas career profile, including work experience, education, tech stack, and job search status.
get_profile	Retrieve your current Himalayas profile information.

Before you enable it

An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.