MCP Server

Tech job intelligence,
inside your AI assistant.

HireJack's Model Context Protocol server gives Claude, Cursor, Cline, and any MCP‑capable client live access to 117,000+ tech jobs, 370+ company hiring profiles, salary benchmarks, resume match scoring, interview prep, and historical market trends — 22 tools across 4 tiers.

claude.ai · Claude Desktop · Claude Code · Cursor · Cline

Connect HireJack to your client

Two transports share one toolset. The hosted HTTP endpoint authenticates you over OAuth and unlocks every tool your HireJack tier permits. The local stdio package runs as a subprocess with the 5 public tools — no account required.

claude.ai (web) All tiers

Open Claude's Customize panel → Connectors → Add custom connector. Paste the endpoint, complete the OAuth flow with your HireJack login, and you're connected.

# Connector URL https://hirejack.com/api/mcp

OAuth 2.1 + PKCE. Your HireJack tier (Free / Pro / Premium / Analyst) determines which tools are visible.

Claude Desktop & Code Public tools

Add the npm package as an MCP server. Runs locally as a subprocess (no auth) and exposes the 5 public tools immediately.

# claude_desktop_config.json { "mcpServers": { "hirejack": { "command": "npx", "args": ["-y", "@hirejack/mcp"] } } }
# Claude Code one-liner claude mcp add hirejack -- npx -y @hirejack/mcp

Cursor, Cline & others Public tools

Any MCP client that supports stdio servers can use the same npm package. Cursor & Cline both honor the standard mcpServers config shape.

# Cursor: ~/.cursor/mcp.json # Cline: settings.json { "mcpServers": { "hirejack": { "command": "npx", "args": ["-y", "@hirejack/mcp"] } } }

Pro+ tools require the OAuth HTTP endpoint — client must support remote MCP servers and OAuth.

22 tools, 4 tiers

Public tools work with the npm package or the hosted endpoint. Pro/Premium/Analyst tools require an authenticated HireJack account on the matching tier and the hosted endpoint.

What you can ask

Drop these prompts into Claude (or any connected client) after setup. The assistant picks the right tool(s) automatically.

Find roles

Find remote senior backend roles with disclosed salary at $180K+.

search_jobs
Research a company

What is Anthropic currently hiring for, and what's their tech stack look like?

get_company_profile
Pro — match scoring

Show me top jobs that match me this week, and tell me which one I should apply to first.

recommendations match_job
Pro — salary

Am I paid well as a senior data engineer in NYC at $215K? What's the percentile?

salary_benchmark
Premium — apply prep

Rewrite my resume bullets for this Stripe role, then give me an interview prep brief.

resume_rewrite interview_prep
Analyst — market research

Which defense tech companies are scaling hiring fastest, and what skills are they prioritizing?

find_breakout_companies find_companies

The full tool reference

Each tool's description and parameter schema match exactly what an MCP client sees when it calls tools/list.

Public tools

No auth required 5 tools · stdio or HTTP

Free for anyone — available via the local npm package or the hosted endpoint. Backed by an embedded daily snapshot so they answer fast and never hit rate limits.

search_jobs

· live jobs · filter + rank

Search HireJack's database of live tech job postings. Filter by keyword, role family, seniority, skill, location, salary, remote policy, or visa sponsorship. Returns a slim list with title, company, location, salary range, posted date, and key skills.

ParamTypeWhat it does
qstringoptFree-text keyword across title, company, domain, and location. Does not search descriptions — use skill for that.
familyenumoptRole family. One of: software_engineering, data_engineering, machine_learning, security, product, design, operations, marketing, sales, finance, people, services, legal, other.
seniorityenumoptintern, junior, mid, senior, staff, principal, manager, director, vp
skillstringoptSingle skill to require (case-insensitive substring), e.g. Rust or Kubernetes.
companystringoptCompany domain filter, e.g. stripe.com.
locationstringoptLocation substring, e.g. San Francisco.
remoteenumoptremote, hybrid, or remote+hybrid.
salary_minnumberoptAnnual USD floor. Matches jobs whose disclosed range could pay at least this much.
salary_maxnumberoptAnnual USD ceiling. Matches jobs whose floor is at or below this.
has_salarybooleanoptOnly return jobs with a disclosed salary.
visaenumoptyes or no.
limitnumberoptMax results (default 25, max 100).
Try saying

"Find me remote senior security engineering jobs that disclose salary and pay at least $220K."

get_job

· one posting in detail

Fetch one job by domain + jobId, or by HireJack detail URL. Returns title, company, location, salary, skills, seniority, posted date, visa/remote/education metadata, and (when available) the AI summary. Use after search_jobs when the user picks a specific role.

ParamTypeWhat it does
domainstringoptCompany domain. Required unless url is provided.
jobIdstringoptCanonical id (exactly as returned by search_jobs). Required unless url is provided.
urlstringoptFull HireJack detail URL, e.g. https://hirejack.com/jobs/stripe.com/sw-eng-12345/.
Try saying

"Pull up the full posting for this URL: https://hirejack.com/jobs/anthropic.com/research-engineer-..."

get_company_profile

· one company · tech stack & trends

Fetch the full hiring profile for one company: tech stack (top skills with counts), role distribution, seniority breakdown, location breakdown, hiring trend (% MoM), salary medians, and an AI‑generated analyst brief.

ParamTypeWhat it does
domainstringreqApex domain (e.g. stripe.com, anthropic.com) — not a subdomain.
Try saying

"What is Stripe hiring for right now, and how is their headcount trending?"

search_companies

· list companies · filter by industry

List HireJack's tracked companies with hiring volume. Filter by industry substring. Useful to discover domains before calling get_company_profile, or to answer "which fintech companies are hiring?" Served from an embedded daily snapshot (see fetchedAt in the response).

ParamTypeWhat it does
industrystringoptSubstring filter (case-insensitive). Examples: fintech, health tech, gaming, defense, AI.
limitnumberoptMax companies (default 50, max 200).
Try saying

"List all the fintech companies you track and rank them by open job count."

get_market_pulse

· market-wide snapshot

Market-wide snapshot: total open tech jobs, tracked companies, remote share, top skills by demand, top companies by hiring volume, and trending skills (week-over-week movers). High-level questions like "how is the tech job market right now?"

No parameters.

Try saying

"Give me a snapshot of the tech hiring market — total jobs, remote share, and what skills are trending up this week."

Pro tools

Pro tier · OAuth required 8 tools · hosted endpoint only

Personal intelligence layered over the market data. Requires an authenticated HireJack user on Pro or higher. Connect via the hosted endpoint and complete the OAuth flow.

match_job

· score a specific posting

Score how well you match a specific job. Returns matchPct (0-100), 5-dimension breakdown (skills 50%, role 20%, seniority 10%, location 10%, remote 10%), matched/missing/bonus skill lists, ATS-specific resume tips for the company's ATS, and a priorityScore that factors hiring velocity.

ParamTypeWhat it does
domainstringreqCompany domain, e.g. stripe.com.
jobIdstringreqThe id field from a search_jobs result, verbatim — do not reformat. Postings rotate; always fetch a fresh id first.
Try saying

"How well do I match this Anthropic research engineer role? Should I apply?"

company_fit

· score your fit for a company

Score how well you fit a company. Returns fitScore (0-100), 5-dimension breakdown (tech stack 40%, roles 25%, seniority 15%, location 10%, hiring trend 10%), matching skills, skill gaps, role alignment, and a hiring-momentum signal.

ParamTypeWhat it does
domainstringreqCompany domain, e.g. anthropic.com.
Try saying

"Am I a good fit for Anthropic? Compare it against Stripe and Cloudflare while you're at it."

recommendations

· top-N jobs scored for you

Top job recommendations for the authenticated user, scored against your profile (skills, desired roles, seniority, location, remote preference). Returns up to ~30 jobs ranked by composite match with per-job details.

ParamTypeWhat it does
limitnumberoptMax recommendations (default 30, max 50).
Try saying

"Show me my top 10 matches this week and group them by which I'm most likely to land."

salary_benchmark

· percentile + career ladder

Compare a salary against the live HireJack market for a role family + seniority. Returns precomputed P10/P25/P50/P75/P90 percentiles, your percentile ranking when salary is supplied, and a career-ladder progression of medians by seniority.

ParamTypeWhat it does
familyenumoptRole family to benchmark against. Same enum as search_jobs.family.
seniorityenumoptSeniority level. Same enum as search_jobs.seniority.
salarynumberoptYour current annual USD salary. When provided, response includes your percentile within the slice.
Try saying

"I make $215K as a senior data engineer. Where does that put me percentile-wise, and what's the staff-level median?"

skill_gap

· what to learn for your target roles

Analyze the gap between your current skills and what your desired roles require in the live market. Returns matchPct, the skills you already have that map to target roles, and the missing skills ranked by impact (how often each skill appears in target-role postings).

No parameters — reads your profile.

Try saying

"What's the skill gap between me and the senior PM roles I'm targeting? What should I learn next?"

market_position

· your overall market score (0-100)

Compute your market position score (0-100). Weighted: 50% skill demand, 20% seniority fit, 15% remote availability for your preference, 15% skill breadth. Returns score, matching jobs count, salary median for matching roles, per-skill demand breakdown, and the top companies whose tech stacks align with you.

No parameters — reads your profile.

Try saying

"What's my market position right now? Which of my skills are in highest demand?"

skill_impact

· simulate adding each missing skill

For each skill you don't have, simulate adding it and compute the market impact. Returns a ranked list with: newJobsUnlocked (jobs that newly match), boostedJobs (matches that would score higher), companiesNeedingIt, relevantRoles, and a composite impactScore. Also returns profileCompleteness.

No parameters — reads your profile.

Try saying

"Which single skill would unlock the most new jobs for me if I added it to my profile?"

watchlist_intelligence

· what's happening at companies you watch

Aggregate hiring intelligence on every company you're watching. Per company: totalJobs, hiringTrend (up/down/stable), trendPct (WoW %), topSkills, medianSalary, sparkline data (~12 weeks), and jobDelta since the previous snapshot.

No parameters — reads your watchlist.

Try saying

"Which of my watched companies are scaling hiring up right now, and which are slowing down?"

Premium tools

Premium tier · OAuth required 2 tools · hosted endpoint only

AI-generated, per-job tailoring. Uses Bedrock Claude under the hood and reads your uploaded resume + the live job description as context. Requires a HireJack Premium subscription.

resume_rewrite

· targeted bullet rewrites

Generate targeted resume bullet rewrites for a specific job. Uses your uploaded resume + the JD as context. Returns before/after bullet rewrites, missing keywords from the JD, and ATS-format tips. You must have a resume uploaded to HireJack (via Settings or Onboarding).

ParamTypeWhat it does
domainstringreqCompany domain, e.g. stripe.com.
jobIdstringreqCanonical job id from search_jobs, verbatim.
Try saying

"Rewrite my resume bullets for this Stripe staff backend role — emphasize what they actually want."

interview_prep

· topics, questions & company research

Generate targeted interview prep for a specific job. Returns: 5 key topics (deep vs surface depth, scaled to seniority), 5 likely questions by type (technical, behavioral, system design) with answer tips, and 3 company-specific research items. Uses JD + company tech stack + your skills as context.

ParamTypeWhat it does
domainstringreqCompany domain, e.g. anthropic.com.
jobIdstringreqCanonical job id from search_jobs results.
Try saying

"Help me prep for my Anthropic interview — what topics will they dig into and what's likely to come up in the system design round?"

Analyst tools

Analyst tier · OAuth required 7 tools · hosted endpoint only

Historical time-series and multi-axis segmentation for investors, journalists, HR researchers, and recruiters in research mode. Daily market history back to March 2026; monthly per-company & per-skill snapshots. Requires the standalone Analyst tier ($49/mo or $399/yr).

get_company_history

· time-series for one company

Monthly snapshots of a company's hiring profile: total jobs, engineering jobs, median salary, hiring trend, top skills, seniority + family distribution. Plus a wider monthlyPostings job-count series and current-state summary.

ParamTypeWhat it does
domainstringreqCompany domain.
monthsnumberoptMonths of history (default 12, max 24). Detailed monthly snapshots began March 2026.
Try saying

"How has Stripe's hiring shifted over the last 12 months? Are they growing or shrinking?"

get_skill_history

· time-series for one skill

Monthly companyCount + jobMentions for a skill, top companies hiring for it each month, and computed MoM deltas.

ParamTypeWhat it does
skillstringreqSkill name or id, e.g. Rust, kubernetes, Machine Learning. Aliases resolve via the skills dictionary.
monthsnumberoptMonths of history (default 12, max 24).
Try saying

"How fast is Rust adoption growing across tech employers? Which companies lead it?"

get_market_history

· market-wide time-series

Time-series of market-wide hiring stats: total jobs, remote share, compensation, seniority + family distribution, top skills. Daily resolution back to early March 2026; monthly resolution from March 2026.

ParamTypeWhat it does
rangeenumoptdaily (up to 90 days back) or monthly (up to 24 months back). Default daily.
daysnumberoptDays of daily history (default 30, max 90; range=daily only).
monthsnumberoptMonths of monthly history (default 12, max 24; range=monthly only).
Try saying

"How has the remote share of tech hiring moved over the last 90 days?"

compare_companies

· up to 5 side-by-side

Compare hiring profiles of multiple companies side-by-side. Per company: current state (totalJobs, medianSalary, trend, top skills) plus monthly snapshot history for the chosen window.

ParamTypeWhat it does
domainsstring[]reqUp to 5 domains, e.g. ["stripe.com", "plaid.com", "adyen.com"].
monthsnumberoptMonths of history per company (default 6, max 12).
Try saying

"Compare hiring at Stripe, Plaid, and Adyen over the last 3 months. Which is growing fastest?"

find_companies

· multi-axis segmentation

Multi-axis segmentation: filter by industry × role family × skill × hiring trend × job-count range. Returns companies sorted by open jobs descending.

ParamTypeWhat it does
industrystringoptIndustry substring filter. E.g. fintech, defense, AI.
familyenumoptCompanies hiring for this role family.
skillstringoptCompanies whose top skills include this (substring match).
trendenumoptup, down, or stable.
minJobsnumberoptMinimum total open jobs (default 0).
maxJobsnumberoptMaximum total open jobs.
limitnumberoptMax companies (default 25, max 100).
Try saying

"Find me defense tech companies with 100+ open engineering jobs and a positive hiring trend."

find_breakout_companies

· extreme hiring growth

Companies with extreme hiring growth right now. Returns companies whose computed trendPct exceeds the threshold, sorted by growth descending, with their recent monthly job counts. Useful for investors and recruiters tracking momentum.

ParamTypeWhat it does
minPctnumberoptMinimum % growth over the trailing window (default 50).
minJobsnumberoptMinimum total open jobs to be considered (default 20; filters out tiny companies).
limitnumberoptMax companies (default 20, max 50).
Try saying

"Show me companies that have doubled their hiring in the last 60 days — filter to ones with 50+ open roles."

find_emerging_skills

· early-stage skill signals

Skills with low-but-growing market adoption — early-stage signals. Computes month-over-month growth in companyCount across the most recent two monthly snapshots and filters to skills still under the "mainstream" threshold.

ParamTypeWhat it does
adoptionMinnumberoptMinimum companies currently mentioning the skill (default 3). Lower = catches earlier-stage skills.
adoptionMaxnumberoptMax companies (default 30) — excludes already-mainstream skills.
growthMinnumberoptMinimum MoM growth % (default 25).
limitnumberoptMax skills (default 20, max 50).
Try saying

"What skills are quietly trending right now — not yet mainstream, but adoption is growing fast?"

What this MCP can — and can't — do

The honest version. HireJack MCP is built for a specific job: tech hiring intelligence. It's not a general-purpose job board, a scraper, or an applicant tracking system.

What it does well

  • Search and filter 117,000+ live tech jobs from 370+ companies — refreshed daily at ~06:35 UTC.
  • Per-company hiring profiles: tech stack, seniority mix, salary medians, hiring trend, AI analyst brief.
  • Match scoring against your actual profile: skills, role family, seniority, location, remote preference. Returns matched / missing / bonus skills and ATS-specific resume tips.
  • Salary benchmarking with real percentiles (P10–P90) per family × seniority slice.
  • Historical trends back to March 2026 — daily market snapshots, monthly per-company & per-skill series.
  • Segmentation queries: industry × family × skill × trend × size, plus breakout-growth and emerging-skill detection.
  • Per-job AI artifacts: resume rewrites and interview prep, generated from your resume + the live JD.
  • Works in every major MCP client: claude.ai (web), Claude Desktop, Claude Code, Cursor, Cline.
  • Always cites a live citation_url so the assistant links to the canonical page.

What it doesn't do

  • Not real-time. Data is refreshed daily — "live" means within the last ~24 hours, not seconds.
  • Tech jobs only. Coverage is engineering, product, design, security, data, ML, ops, and supporting functions at tech-adjacent companies. Not retail, healthcare-clinical, hospitality, etc.
  • Not every company. ~370 tracked companies (and growing). If a company isn't in the seed list, it isn't here.
  • Doesn't submit applications. Every job links to the company's own ATS; the assistant won't apply on your behalf.
  • No PII on other users. The candidate index (for recruiters on hirejack.com) is two-toggle opt-in and not exposed via MCP.
  • Pro+ tools need an account. The local npm package only exposes the 5 public tools. Pro/Premium/Analyst tools require the hosted endpoint + OAuth + matching tier.
  • History is shallow on day 1. Detailed monthly snapshots began March 2026; depth grows month-over-month.
  • No raw resume parsing endpoint. Resume context comes from what you uploaded to HireJack via the website, not from a file you paste into a tool call.

Frequently asked

Do I need a HireJack account?
Only if you want the Pro, Premium, or Analyst tools. The 5 public tools (search_jobs, get_job, get_company_profile, search_companies, get_market_pulse) work without an account — just install the npm package. Everything else requires an OAuth sign-in via the hosted endpoint.
Which transport should I use?
If your client supports remote MCP servers (claude.ai web, Claude Desktop with remote support, Cline with remote support) and you want Pro+ tools: use the hosted endpoint https://hirejack.com/api/mcp. Otherwise the @hirejack/mcp npm package gives you the public tools as a local stdio subprocess.
How fresh is the data?
Refreshed daily at ~06:35 UTC. Search and company profiles reflect what was on company career sites as of the most recent run. Historical snapshots (monthly company / skill profiles, daily market stats) accumulate from March 2026 onward.
Does the npm package phone home or track me?
No. The npm package is a local stdio subprocess that hits HireJack's public APIs with no user identifier. The hosted endpoint logs per-tool usage analytics (tool name, duration, response size, your user id when authenticated) so we can keep the service running and improve coverage — same as any SaaS API. We don't log your prompts or your tool arguments verbatim.
What's the package on npm?
@hirejack/mcp — installable via npx -y @hirejack/mcp. Also listed on the official MCP registry as io.github.TVLSS/hirejack-mcp. Source for the stdio bundle is open at TVLSS/hirejack-mcp.
My client doesn't see any tools after I connect.
Three things to check. 1. If you're using the hosted endpoint, complete the OAuth flow (the connector popup logs you into HireJack). 2. If you only see 5 tools, your tier is on Free (or your token didn't authenticate) — upgrade in pricing or check that your hj_token is set in localStorage. 3. Restart the client after editing the config; stdio servers don't hot-reload.
Can I see the source?
Yes — the stdio bundle (5 public tools) is open-source at github.com/TVLSS/hirejack-mcp. The hosted HTTP transport (Pro+ surface) runs on Lambda against HireJack's internal data layer and isn't published — same business reason job boards keep their indexes private.
How do I report a bug or request a tool?
Drop a note through the feedback form — tag it "feature request" and mention MCP. We read every submission.

Get HireJack talking to your AI in 60 seconds.

Paste https://hirejack.com/api/mcp into your client's custom connector field, or npx -y @hirejack/mcp into your stdio config. Sign up for Pro to unlock match scoring, skill impact, and the rest.