English Français العربية

io.github.erashev-fedor/gemsift-mcp

Official registry Productivity

AI resume triage for recruiters. Query your candidate pool from Claude or ChatGPT.

Source code Documentation

What can you do with io.github.erashev-fedor/gemsift-mcp?

  • "Show me the shortlist for my Backend Engineer role, and flag any hidden gems."
  • "Why is Maria ranked above the other candidates for this position?"
  • "Any duplicate or suspicious applications in my pool right now?"
  • "Create a role from this job description and tell me who already fits."
  • "Move James to shortlist and pass on the rest for the Designer role."

Install

Add this to your client's MCP configuration (Claude Desktop: claude_desktop_config.json; Cursor: .cursor/mcp.json; VS Code and others use the same "mcpServers" shape).

Remote (streamable-http)
Claude Code 
claude mcp add --transport http gemsift-mcp 'https://gemsift.com/api/mcp/mcp' --header 'Authorization: <Bearer <your per-workspace key from gemsift.com -> Settings -> Connect AI>>'
JSON configuration (Claude Desktop, Cursor, VS Code) 
{
    "mcpServers": {
        "gemsift-mcp": {
            "type": "http",
            "url": "https://gemsift.com/api/mcp/mcp",
            "headers": {
                "Authorization": "<Bearer <your per-workspace key from gemsift.com -> Settings -> Connect AI>>"
            }
        }
    }
}

Documentation

From the project's README on GitHub, refreshed weekly. Links open on GitHub.

Contents
  1. Gemsift MCP Server
  2. Who it's for
  3. What you can do (ask your AI things like)
  4. Connect
  5. Tools
  6. How it is different

Gemsift MCP Server

Query your recruiting pipeline from Claude or ChatGPT.

Gemsift is an AI resume-triage tool for recruiters. This MCP server connects your AI assistant directly to your own candidate pool, so you can screen, shortlist, and move candidates by chat. Drop in a pile of applicants and ask your AI who is actually worth a call, including the strong people a keyword filter would normally bury, with the reasons why. The recruiter always makes the final call.

Who it's for

Individual and agency recruiters, and small in-house hiring teams who do their own first-pass screening, drown in inbound applications, and do not want to lose a good candidate just because they look weak on paper.

What you can do (ask your AI things like)

  • "Show me the shortlist for my Backend Engineer role, and flag any hidden gems."
  • "Why is Maria ranked above the other candidates for this position?"
  • "Any duplicate or suspicious applications in my pool right now?"
  • "Create a role from this job description and tell me who already fits."
  • "Move James to shortlist and pass on the rest for the Designer role."

Connect

Hosted, remote server. No install. You authenticate with a per-workspace key from your Gemsift account: gemsift.com → Settings → Connect AI.

  • Streamable HTTP: https://gemsift.com/api/mcp/mcp
  • SSE (legacy): https://gemsift.com/api/mcp/sse
  • Auth header: Authorization: Bearer <your-workspace-key>
{
  "mcpServers": {
    "gemsift": {
      "url": "https://gemsift.com/api/mcp/mcp",
      "headers": { "Authorization": "Bearer <your-workspace-key>" }
    }
  }
}

Tools

ToolWhat it does
list_positionsYour open roles with candidate counts
get_shortlistTop candidates for a role, including hidden gems
get_candidate_analysisFull parsed profile and per-role evaluations for one person
why_rankedWhy a candidate scores the way they do: matched, missing, transferable
list_trust_flagsDuplicate, consistency, and fraud signals in your pool
get_email_forwarding_addressA forwarding address to send resumes straight into the pool
create_positionCreate a role from a job description
set_candidate_statusMove a candidate (shortlist, pass, and so on) by chat

How it is different

Most screeners rank against keywords and quietly drop everyone below a threshold. Gemsift does the opposite: it surfaces the strong people who read weak on paper, unifies applicants from every source into one pool, removes duplicates, and flags likely fraud, all with the reasoning visible so you can trust or overrule it.

Learn more and get a key: **https://gemsift.com**

Frequently asked questions

What is io.github.erashev-fedor/gemsift-mcp?

AI resume triage for recruiters. Query your candidate pool from Claude or ChatGPT.

How do I install io.github.erashev-fedor/gemsift-mcp?

It runs as a remote server. In Claude Code: claude mcp add --transport http gemsift-mcp 'https://gemsift.com/api/mcp/mcp' --header 'Authorization: <Bearer <your per-workspace key from gemsift.com -> Settings -> Connect AI>>'. For other clients, paste the JSON configuration from this page into their MCP settings.

Does io.github.erashev-fedor/gemsift-mcp need an API key?

Yes. It expects Authorization, which you set in its configuration.

Is io.github.erashev-fedor/gemsift-mcp open source?

Its code is public on GitHub, but the repository does not declare a standard license.

Who maintains io.github.erashev-fedor/gemsift-mcp?

It is published by erashev-fedor. AI.tn lists it but does not maintain or endorse it.

From the official MCP Registry. Listing is not an endorsement: review a server's code and permissions before giving it access to your data.