Skip to content

Resume parsing and candidate matching MCP server

The HireLayer MCP server lets Claude, ChatGPT, Cursor, VS Code, Codex and any MCP client parse resumes, turn job descriptions into criteria, then score and rank candidates, with an explanation for every result.

URLhttps://hirelayer.co/mcp

Add the URL in Claude or ChatGPT and sign in: no API key to copy. 50 free credits a month, no card required.

Your AI assistant · hirelayer connectedMCP

Screen these 3 resumes against the Senior React job and tell me who to interview.

  1. extract_job_criteria1 job description
  2. parse_resume3 resumes
  3. score_resumes3 resumes, ranked

Here is the shortlist, scored criterion by criterion:

  • Alex MorganStrong match: React, TypeScript, team leadInterview
  • Camille DurandPartial: senior front end, Vue.js rather than ReactMaybe
  • Sam LeeWeak: back-end profile with some ReactPass
Fictional sample files from the repository. This run uses 6 credits: 3 for parsing and 3 for scoring. Extracting the criteria is free.
Server URL
https://hirelayer.co/mcp
Transport
Streamable HTTP, or stdio with npx
Authentication
OAuth sign-in, or API key
Registry
co.hirelayer/hirelayer
Local package
hirelayer-mcp
License
MIT, open source

Six recruiting tools for your AI assistant

Screen applicants in a chat, build a recruiting agent, enrich an ATS or prototype HR tech features without writing integration code. The tools keep resumes, jobs and scores in your HireLayer account, like the REST API, and never change anything in your own systems.

  • parse_resume

    Resume Parser

    Turns a resume file into structured JSON: contact details, work experience, education, languages, skills and the full text, stored under an application_id. Scanned resumes go through OCR, and the resume language is detected.

    Input: An attached file, a public file URL, a base64 file with its name or the resume text; a local file path with the npx server. PDF, DOC, DOCX, ODT, RTF, TXT, HTML, Markdown, PPT, PPTX, ODP, XLS, JPG, PNG or BMP, under 4.5 MB.

  • extract_job_criteria

    Job Criteria Extractor

    Returns the typed criteria of a job: a requirement, a weight from 0 to 3 and a threshold for each. The job is stored under a job_id. Free.

    Input: Job description text, in any language.

  • score_resumes

    Resume Scorer

    Scores 1 to 50 resumes against a job from 0 to 1, best first, with a verdict, a summary and the evidence for each criterion. precise_ranking re-ranks the 10 best with a reason each.

    Input: A job text (with its criteria if you have them) or a job_id, and resume texts or application_ids.

  • search_resumes

    Talent Pool Search

    Finds the stored resumes closest to a job, by proximity only, with a filter on location and date added.

    Input: A job_id or a job text, and how many resumes to return.

  • search_jobs

    Job Recommendations

    Finds the stored open jobs closest to a resume, by proximity only.

    Input: An application_id or a resume text, and how many jobs to return.

  • normalize_skills

    Skills Normalizer

    Maps free-text skills to a skills taxonomy with stable IDs, families and domains, named in French or English.

    Input: Free text, from one skill to a whole skills section.

  • Ready-made prompts

    Clients show these as commands you can run in one click.

    screen_candidates
    Runs the full screening workflow (parsing, scoring with precise ranking) and writes a shortlist.
    summarize_resume
    Parses one resume and writes a recruiter summary.
    normalize_skills_list
    Normalizes a skills section and groups it by domain.

Up and running in three steps

About two minutes from sign-up to your first parsed resume.

  1. STEP 01

    Add the server URL

    In Claude, ChatGPT, Cursor or VS Code, add a server with the URL https://hirelayer.co/mcp, or use a one-click install below.

  2. STEP 02

    Sign in to HireLayer

    Your app opens the HireLayer sign-in page: log in or create a free account (50 credits a month, no card) and allow access. No API key to copy.

  3. STEP 03

    Ask your assistant

    For example: attach a resume and ask “Summarize this candidate.” The assistant picks the right tool and explains the result.

Add HireLayer to your MCP client

Add the server URL, then sign in to HireLayer when your app asks. The local npx server uses an API key instead.

https://hirelayer.co/mcp
https://hirelayer.co/mcp
claude mcp add --transport http hirelayer https://hirelayer.co/mcp
{
  "mcpServers": {
    "hirelayer": { "url": "https://hirelayer.co/mcp" }
  }
}
{
  "servers": {
    "hirelayer": { "type": "http", "url": "https://hirelayer.co/mcp" }
  }
}
codex mcp add hirelayer --url https://hirelayer.co/mcp
codex mcp login hirelayer
{
  "mcpServers": {
    "hirelayer": {
      "command": "npx",
      "args": ["-y", "hirelayer-mcp"],
      "env": { "HIRELAYER_API_KEY": "your-api-key" }
    }
  }
}

Where it goes

Claude
Claude (web and desktop): Settings → Connectors → Add custom connector, paste the URL, then Connect and sign in to HireLayer.
ChatGPT
ChatGPT (web, Business, Enterprise, Edu, or Pro in developer mode): open Plugins → + → Add custom MCP server, paste the URL, choose OAuth, then sign in to HireLayer.
Claude Code
Run the command, then /mcp in Claude Code to sign in.
Cursor
Use the one-click install, or add the server to ~/.cursor/mcp.json, then click Connect.
VS Code
Use the one-click install, or add this to .vscode/mcp.json. VS Code opens the sign-in page.
Codex CLI
Run both commands; the second one opens the sign-in page.
Local (npx)
Any client that runs stdio servers (Claude Desktop, Windsurf, Cline, Zed, LM Studio…): runs on your machine with an API key, and reads local resume files.

To read resume files from your disk, run the server locally with npx -y hirelayer-mcp and the HIRELAYER_API_KEY environment variable.

Step-by-step guides: ChatGPT and Claude.

Ask in plain language

Your assistant chooses the tools, chains them and explains the result.

Recruiters and hiring managers

  • “Parse ~/Downloads/jane-doe.pdf and summarize her experience in five bullet points.”
  • “Here is our job description for a Senior Data Engineer. Extract the criteria, then tell me which ones are must-haves.”
  • “Score the resumes in ~/candidates/ against this job and give me a shortlist table with scores and main gaps.”
  • “Rank these 8 candidates for the Account Executive role and explain why the top 3 stand out.”
  • “Does this candidate meet every mandatory criterion? If not, which ones are missing?”

Developers and HR tech teams

  • “Parse this resume and map the result to our ATS candidate schema: { name, email, current_title, skills[] }.”
  • “Normalize this skills section: MS Office (Word, Excel), React.js, fluent English, project management.”
  • “Write a TypeScript function that sends a resume to the HireLayer API, using the JSON this tool returned as the expected type.”

No resumes at hand? The repository ships a fictional job and three resumes. Try the sample files.

How candidate screening works

Ask for a shortlist and the assistant chains the tools for you. The screen_candidates prompt runs the same workflow in one command.

  1. 01

    Read the job

    extract_job_criteria

    The job description becomes weighted criteria, with must-haves flagged.

  2. 02

    Read each resume

    parse_resume

    Every CV becomes structured JSON and plain text the next tools can score.

  3. 03

    Score criterion by criterion

    score_resumes

    Up to 50 candidates in one call, best first, each with a score from 0 to 1 and an explanation per criterion.

  4. 04

    Then rank the best

    score_resumes + precise_ranking

    The 10 best are re-ranked against each other, each with a reason, at no extra cost.

A reason behind every score

score_resumes returns up to 50 candidates best first, each with a score from 0 to 1, a verdict, a summary and, for every criterion, its weight, whether it is required, a status, the evidence quoted from the resume and an explanation. With precise_ranking, the 10 best are re-ranked against each other, each with a reason.

Criteria, summaries and explanations follow the job’s language when it is English, French, German, Spanish or Portuguese, and are in English otherwise. Your assistant translates them when it answers you in another language.

Scores support human decisions; they don’t replace them.

score_resumes result (excerpt)JSON
{
  "precise_ranking": true,
  "scores": [
    {
      "rank": 1,
      "application_id": "file_2c982f413a5e9e79563f8a1fcb5c7caa",
      "score": 0.86,
      "verdict": "highly_relevant",
      "summary": "Strong fit: React, TypeScript and the experience asked for are shown in recent roles. English level to confirm.",
      "ranking_reason": "The only candidate who has led a React team on a product in production.",
      "required_missing": [],
      "to_verify": [{ "criterion_id": "crit_4", "label": "English" }],
      "criteria": [
        {
          "criterion_id": "crit_1",
          "label": "React",
          "requirement": "required",
          "weight": 3,
          "status": "met",
          "score": 1,
          "confidence": "high",
          "evidence": "Lead Front-End Engineer - Brightline (2022-25): 'Led a team of 5 React developers on a platform in production'",
          "explanation": "Has led a React team on a production platform since 2022."
        }
      ]
    }
  ]
}

Billed like the REST API

The MCP server is free and open source. Tool calls use the credits of your HireLayer plan, shared with the REST API.

  • 1 credit per resume parsed, per search and per skills call; extracting job criteria is free. Resume Scorer counts 1 credit per resume evaluated, plus 1 per resume file parsed, and its precise_ranking option 1 per call; a pair already scored is free.
  • Precise ranking adds 1 credit to a score_resumes call.
  • Failed calls are not charged.
PlanCreditsPrice
Free50 a monthFree, no card required
Starter500 a month€24 / month
Scale2,000 a month€89 / month
EnterpriseCustom volumeOn quote
See plans and credit packs

Sends only what you ask, under your control

  • You choose the apps

    An app gets access only after you sign in and allow it. Disconnect it any time in Dashboard → AI assistants.

  • Only the files you send

    Resumes reach HireLayer over HTTPS only when a tool is called with them. The local server reads only the file paths you name.

  • Storage you can turn off

    A parsed resume is kept like any API parse: the file is stored in Paris, France, and the parsed resume is saved to your HireLayer account. A workspace set with do_not_store_data keeps both out of storage.

  • Humans decide

    Resumes contain personal data: use the tools in line with your hiring process and rules such as GDPR.

Read the privacy policy

HireLayer MCP server FAQ

Comparing options? Our guide lists every recruiting MCP server, from resume parsing to ATS connectors.

Is there a resume parser API with an MCP server for Claude or ChatGPT?

Yes. HireLayer Resume Parser is a resume parsing API with an official MCP server at https://hirelayer.co/mcp. Add that URL in Claude or ChatGPT, sign in to HireLayer, then ask the assistant to parse a resume: its parse_resume tool returns the same JSON as the REST API, with contact details, work experience, education, languages and skills, from a PDF, a Word file, a scan or a photo of a CV. Each resume parsed uses 1 credit, and the Free plan includes 50 credits a month with no card required.

What is the HireLayer MCP server?

It is the official Model Context Protocol server for HireLayer. It gives AI assistants such as Claude, ChatGPT, Cursor, VS Code Copilot and Codex six recruiting tools, one per HireLayer product: Resume Parser, Job Criteria Extractor, Resume Scorer, Talent Pool Search, Job Recommendations and Skills Normalizer. It is open source (MIT) and published on npm as hirelayer-mcp.

Which MCP clients does it work with?

Claude (web, desktop and Claude Code), ChatGPT, Cursor, VS Code with GitHub Copilot, Codex and any client that supports remote MCP servers connect to https://hirelayer.co/mcp and sign in with OAuth. Clients that only run local servers use npx -y hirelayer-mcp with an API key.

Do I need to write code?

No. Once the server is added to your client, you ask in plain language and the assistant calls the tools. Developers can use the same tools to prototype an integration, then call the REST API from their own code.

How much does it cost?

Tool calls use the same credits as the REST API. 1 credit per resume parsed, per search and per skills call; extracting job criteria is free. Resume Scorer counts 1 credit per resume evaluated, plus 1 per resume file parsed, and its precise_ranking option 1 per call; a pair already scored is free. Failed calls are not charged. The Free plan includes 50 credits a month with no card required. Paid plans add more credits.

Is HireLayer an ATS?

No. HireLayer provides the AI building blocks of recruiting software: resume parsing, matching, ranking and skills. Use them on their own through MCP, or plug them into your ATS or HR tech product through the REST API.

Which language are the results in?

The text that HireLayer writes (criteria, summaries, explanations and ranking reasons) follows the job’s language when it is English, French, German, Spanish or Portuguese, and is in English otherwise; the evidence quotes the resume as written. Your assistant translates it when it answers in another language. normalize_skills returns French or English labels.

Are my resumes stored?

Resumes are sent to the HireLayer API over HTTPS only when a tool is called. Like any API parse, the file is then stored in Paris, France, and the parsed resume is saved to your HireLayer account. A workspace set with do_not_store_data keeps both out of storage; the local npx server can also set the option to true on a call.

Can I keep each client’s candidates apart?

Yes, with the hosted server. If you recruit for several client companies, give the client’s workspace id when you ask (letters, digits and . _ : ~ -, up to 200 characters, such as acme-corp): the tools keep that client’s resumes, jobs and scores in its workspace, apart from the other clients’ candidates. Without one, they go to your default workspace. Clients that let you add headers can also send HireLayer-Workspace with the connection to change that default, for example claude mcp add --transport http hirelayer https://hirelayer.co/mcp --header "HireLayer-Workspace: acme-corp".

Do I need an API key?

No, with the hosted server at https://hirelayer.co/mcp: you sign in to HireLayer and allow access, and the app's calls are billed to your plan. You can disconnect an app at any time in Dashboard → AI assistants. The local npx server uses an API key instead.

Can I use the REST API instead?

Yes. The MCP server calls the same public API. For code integrations, call the REST API directly: see the API reference, the OpenAPI document and llms.txt.

Give your assistant a recruiting toolkit today.

Create a free account, add the server to your client and sign in. Building it into your own product? Call the same APIs over REST.