Candidate Screener MCP Server
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
π― AI Candidate Screener β Model Context Protocol (MCP) Server
An autonomous, evidence-driven Model Context Protocol (MCP) Server for technical recruitment and ATS resume screening (default: Senior SDET / QA Automation, extensible to any Job Description).
Connect this MCP server to Google Antigravity IDE, VS Code (GitHub Copilot / Cline / Cursor / Roo-Code), or Claude Desktop to empower your AI assistant to parse multi-format resumes, eliminate keyword inflation, and generate audit-ready gap matrices.
β¨ Features & Capabilities
Strict "No-Skill-Inflation" 4-Tier Verification Engine:
π’ Matched (1.00Γ weight): Verified in $\ge 1$ concrete project deliverable.
π Partial Match (0.60Γ weight): Adjacent tech or minimal exposure.
π΅ Claimed not evidenced (0.30Γ weight): Keyword list only, no proof.
π΄ Missing (0.00Γ weight): Not found in resume.
100-Point Capacity Rubric: Evaluates candidates across Mandatory Skills (85 pts), Good-to-Have Bonus (10 pts), and Experience Fit (5 pts).
Hard-Fail Rule Overrides: Automatically catches core language gaps and AI/Agentic testing gaps.
Multi-Transport Support: Runs as a local
stdioserver or remotesse(Server-Sent Events) HTTP service.Cross-Platform Compatibility: Works on Windows, macOS, Linux, and Cloud (Render, Docker, Cloudflare, Hugging Face).
Related MCP server: Batcave-MCP
π οΈ MCP Tools, Resources & Prompts
π§° Tools
Tool Name | Parameters | Description |
|
| Evaluates a single candidate resume against the active JD. Returns 100-pt score breakdown, Gap Matrix, red flags, and final verdict. |
|
| Batch screens multiple profiles and produces a ranked leaderboard. |
| None | Retrieves the active ground-truth Job Description requirements. |
| None | Retrieves the 100-point capacity scoring model and weight matrix. |
π Resources
URI | Description |
| The active ground-truth Job Description requirements. |
| The active 100-point scoring model and weight allocations. |
π¬ Prompts
Prompt Name | Parameters | Description |
|
| Generates a structured prompt instructing the LLM to screen a candidate using the 4-tier engine. |
π Quick Setup Guide
1. Installation
Clone this repository and install dependencies:
git clone https://github.com/surendra1220/candidate-screener-mcp.git
cd candidate-screener-mcp
pip install -r requirements.txtπ How to Add to Your AI Tools
A. Google Antigravity IDE
Add to your Antigravity configuration file (~/.gemini/config/mcp_config.json):
{
"mcpServers": {
"candidate-screener": {
"command": "python",
"args": [
"/path/to/candidate-screener-mcp/mcp_server.py"
]
}
}
}(On Windows, use py with args ["-3", "C:\\path\\to\\candidate-screener-mcp\\mcp_server.py"])
Alternatively, in Antigravity IDE:
Click Additional Options (
...) in the top right.Select MCP Servers $\rightarrow$ Add Server.
Paste the configuration above.
B. Visual Studio Code (Copilot Agent Mode / Cline / Roo-Code / Cursor)
Add to your project's .vscode/mcp.json:
{
"mcpServers": {
"candidate-screener": {
"command": "python",
"args": [
"${workspaceFolder}/mcp_server.py"
]
}
}
}C. Claude Desktop
Add to your Claude Desktop configuration file:
macOS:
~/Library/Application Support/Claude/claude_desktop_config.jsonWindows:
%APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"candidate-screener": {
"command": "python",
"args": [
"C:\\path\\to\\candidate-screener-mcp\\mcp_server.py"
]
}
}
}π Running as a Public / Remote SSE Server
You can also host the MCP server as a public HTTP SSE endpoint for team use:
Run Locally / Cloud Server:
python mcp_server.py --transport sse --host 0.0.0.0 --port 8000Connect via Remote SSE in any MCP Client:
{
"mcpServers": {
"candidate-screener-remote": {
"serverUrl": "https://your-domain.com/sse"
}
}
}π‘ Example Prompts to Ask Your AI Assistant
Once connected, you can interact with the Candidate Screener naturally:
Screen an uploaded candidate:
"Using candidate-screener, screen the attached resume against our default SDET job description."
Screen with a custom role:
"Evaluate this candidate against the custom JD provided in this prompt using the 4-tier no-inflation engine."
Inspect the active rubric:
"What are the mandatory skill weights and AI hard-fail override rules in the active rubric?"
π¦ Project Structure
candidate-screener-mcp/
βββ mcp_server.py # Main FastMCP Server (Tools, Resources, Prompts)
βββ requirements.txt # Python dependencies (mcp, pypdf, python-docx, fpdf2)
βββ references/
β βββ job-description.md # Ground-truth SDET Job Description
β βββ rubric.md # 100-point capacity scoring model & rules
βββ .vscode/
β βββ mcp.json # VS Code MCP configuration
βββ antigravity_mcp_config.json # Antigravity IDE configuration
βββ Dockerfile # Container deployment configuration
βββ README.md # Public documentationπ License
This project is open source and available under the MIT License.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelogβ¦
This server cannot be deployed
Maintenance
Related MCP Connectors
Recruiter-style resume checker: scores, shortlist decisions, risks, CV rewrites and interview prep
AI resume triage for recruiters. Query your candidate pool from Claude or ChatGPT.
AI screening interviews, resume matching, and evidence-linked scorecards for recruiting teams.
AI interview screening: create interviews from a JD, invite candidates, read evidence-backed scores.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceEnables automatic analysis and comparison of CVs against a job description, scoring candidates and generating professional reports.-
- AlicenseNot gradedqualityDmaintenanceEnables resume review against job descriptions through a three-stage pipeline: match scoring, experience rewriting, and ATS optimization. Supports text or document uploads.5 npmMIT
- FlicenseNot gradedqualityCmaintenanceEnables AI agents to upload resumes and receive structured ATS scores with parseability, section coverage, contact-info, and keyword analysis, along with qualitative improvement suggestions via an LLM.-
- AlicenseAqualityBmaintenanceTurns any AI assistant into an ATS-optimized resume engine that tailors resumes to job descriptions, validates candidate intake, analyzes gaps, scores matches, and renders DOCX/PDF outputs.9MIT