# HireLayer [Health: Active]

**Category:** 🏢 Workplace & Productivity  
**Repository:** https://github.com/HireLayer/hirelayer-mcp  
**GitHub Stars:** 0  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/hirelayer

## Description
Remote MCP server (Streamable HTTP, OAuth 2.1) for recruiting: parse_resume turns PDF, Word or scanned CVs into structured JSON, extract_job_criteria turns a job description into weighted criteria, match_candidate and rank_candidates score candidates against a job with an explanation per criterion, and resolve_skills normalizes skills to a taxonomy. All tools are read-only. Also available as the npm package hirelayer-mcp. Free plan with 50 credits a month, no card required.

## Tools
Capabilities this server exposes over MCP:

- **parse_resume** — Parses a resume or CV file into structured JSON: contact details, experience, education, languages, skills and the full text
- **extract_job_criteria** — Turns a job description into weighted criteria: a weight from 1 to 3, a mandatory flag and a rationale for each
- **match_candidate** — Scores one candidate against a job from 0 to 1, with a summary and a status and explanation for each criterion
- **rank_candidates** — Ranks up to 10 candidates for one job, with a score and a rationale for each
- **resolve_skills** — Maps free-text skills in French or English to taxonomy skills, with their families and domains

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "hirelayer": {
    "command": "npx",
    "args": ["-y","hirelayer-mcp"],
    "env": {
      "HIRELAYER_API_KEY": ""
    }
  }
}
```

**Requires environment variables:** `HIRELAYER_API_KEY` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What HireLayer MCP server does

HireLayer MCP server exposes read-only recruiting operations to MCP-compatible AI clients. It connects to HireLayer and provides a sequence of tools for turning unstructured hiring documents into comparable candidate data.

`parse_resume` accepts supported resume or CV files and returns structured JSON containing contact details, work experience, education, languages, skills, and the document’s full text. Scanned documents can use OCR, and the resume language is detected. `extract_job_criteria` converts a job description into criteria with weights from 1 to 3, mandatory flags, and rationales.

For comparison, `match_candidate` scores one candidate against a job on a 0-to-1 scale and explains the result for each criterion. `rank_candidates` orders up to 10 candidates for one job and returns a score and rationale for each. `resolve_skills` maps free-form French or English skills to taxonomy entries, including their families and domains.

## How it works

The hosted endpoint is `https://hirelayer.co/mcp`. Clients authenticate by signing in through OAuth 2.1, so the hosted setup does not require copying an API key. Clients that cannot complete OAuth can send a HireLayer API key as a bearer token.

A typical workflow starts with a resume and job description. The agent can parse the resume, derive weighted job criteria, then match one candidate or rank a group. The tools are read-only: they analyze supplied recruiting data but do not modify an applicant-tracking system or other external records.

## Setup and configuration

HireLayer MCP server can be used as a hosted Streamable HTTP server or installed locally through the `hirelayer-mcp` npm package. For the hosted version, add the server URL to an MCP client and sign in when prompted. The README documents setup for Claude, ChatGPT, Claude Code, Cursor, VS Code, Codex CLI, and other clients.

For local stdio use, run `npx -y hirelayer-mcp` and provide the `HIRELAYER_API_KEY` environment variable. This option is suitable when the agent needs to read a file from the local machine. Docker is also documented; when running in Docker, local files must be mounted into the container, or `file_url` must be supplied.

The hosted parser can receive a file attached in ChatGPT, a public HTTPS URL, a small base64-encoded file, or resume text. The documented file formats include PDF, DOC, DOCX, ODT, RTF, TXT, PPT, PPTX, ODP, XLS, JPG, PNG, and BMP, with a hosted file-size limit below 4.5 MB.

## Tools and capabilities

- Parse resumes and CVs into structured candidate fields and full text.
- Extract weighted, explained job criteria and mandatory requirements.
- Score one candidate against a job with criterion-level explanations.
- Rank as many as 10 candidates for one job.
- Normalize French and English skills to taxonomy skills, families, and domains.

## Limitations and notes

HireLayer MCP server provides read-only tools, so it is intended for analysis, screening, enrichment, or prototyping rather than writing hiring decisions back to another system. The hosted service requires HireLayer authentication and uses account credits; the README states that the free plan includes 50 credits per month with no card required.

For local file parsing, use the local package or mount files into Docker. The hosted endpoint instead expects an attachment, URL, base64 data, or text. Candidate scores and rankings are outputs for an agent workflow; the supplied material does not describe automated hiring actions or persistence of results.

_Full upstream README: https://allmcps.com/mcp/hirelayer/readme_