openingsby Avagama
Free to explore. No sign-up needed.

Good opportunities.
A simpler way to find them.

Search job postings across companies. Filter what matters.
Open a role when you’re ready to take the next step.

Explore recent roles in India

View all roles →

Recently added to the index. Search above to narrow these to your interests.

–roles posted in the last 7 days
–posted in the last 30 days
–open to India, last 7 days
  • Loading recent roles…

Checking the shared job index…

Go beyond keywords. Find your fit.Connect Claude, ChatGPT, or your AI app for evidence-based resume matching.Set up AI ↗

Add it to your AI app

One address, any app. Paste it where your app takes a custom connector or remote MCP server, sign in with the code we email you, and start a chat. Nothing to install.

connector address
https://openings.avagama.co/mcp

Claudeweb, desktop, mobile

Free, Pro, Max, Team, and Enterprise. Free accounts get one custom connector.

  1. Open the connector form

    Customize → Connectors → + → Add custom connector.

  2. Fill it exactly like this, then Add

    Name
    Openings
    URL
    https://openings.avagama.co/mcp
    Authentication
    Always required Detected
    OAuth client
    Use Anthropic's hosted client metadata Recommended · Detected
    Request headers
    none
    Advanced
    leave closed

    If "Detected" sits next to a different OAuth option, pick that one. Both work.

  3. Sign in

    Claude opens our page. Enter your email, type the six-digit code from the email, click Allow.

  4. Let the tools run

    Back on the connector page, set Tool permissions to Always allow. Every Openings tool is read-only; nothing can apply, submit, or write on your behalf, so approving each call one by one only slows the conversation.

  5. Start a chat

    Click + at the lower left → Connectors → switch Openings on. Paste the starter prompt.

Team and Enterprise: an owner adds it first under Organization settings → Connectors → Add → Custom → Web. Members then click Connect.

ChatGPTweb

Needs Developer mode. Plus, Pro, Business, Enterprise, and Education on the web.

  1. Turn on Developer mode

    Settings → Security and login → Developer mode.

  2. Create the connector

    Settings → Connectors (also listed as Apps) → Create.

    Name
    Openings
    MCP server URL
    https://openings.avagama.co/mcp
    Authentication
    OAuth
    Client ID and secret
    leave empty; ChatGPT identifies itself to us automatically

  3. Sign in

    ChatGPT opens our page. Enter your email, type the six-digit code, click Allow.

  4. Start a chat

    Open the + menu → Developer mode → tick Openings. Paste the starter prompt.

Claude Codeterminal

  1. Add the server once, for every project

    claude mcp add --transport http --scope user openings https://openings.avagama.co/mcp

  2. Sign in

    Start Claude Code and type /mcp, choose Openings, and follow the browser sign-in. Newer versions also accept claude mcp login openings from the terminal.

Codex CLIterminal

  1. Add and sign in

    codex mcp add openings --url https://openings.avagama.co/mcp
codex mcp login openings

  2. Or edit the config by hand

    The server lands in ~/.codex/config.toml under [mcp_servers.openings] with url = "https://openings.avagama.co/mcp".

Cursoreditor

  1. Add the server

    Put this in ~/.cursor/mcp.json for every project, or in .cursor/mcp.json inside one project.

    {
  "mcpServers": {
    "openings": { "url": "https://openings.avagama.co/mcp" }
  }
}

  2. Sign in

    Cursor opens the sign-in the first time you enable the server or use one of its tools in chat.

VS Codewith Copilot

  1. Add the server

    Run MCP: Add Server from the Command Palette and choose HTTP, or put this in your user mcp.json (MCP: Open User Configuration) or in .vscode/mcp.json.

    {
  "servers": {
    "openings": { "type": "http", "url": "https://openings.avagama.co/mcp" }
  }
}

  2. Sign in and pick tools

    VS Code prompts for the sign-in when the server starts. In Chat, Configure Tools lists the seven Openings tools.

Gemini CLIterminal

  1. Add the server

    gemini mcp add --transport http openings https://openings.avagama.co/mcp

  2. Sign in

    Inside Gemini CLI run /mcp auth openings to complete the email sign-in.

Any MCP app

Add a remote or HTTP MCP server with the address https://openings.avagama.co/mcp. Openings speaks standard OAuth with automatic client registration and client metadata documents, so no client ID or secret is needed. If your app insists on one, write to [email protected] and we will issue it.

Then paste this to start

Show me what Openings covers in India. If that looks useful, ask for my resume and use it to find backend roles, including good jobs whose titles I would not have searched for. Rank by evidence and explain every gap.

Your assistant reports how many employers and roles it holds for your country, asks for your resume, then returns roles posted in the last 7 days ranked with quoted evidence. It widens to 14 and 30 days only when a week is thin, and says so.

Prefer to run it on your own machine?

The same tools, fully local: your resume never leaves your computer and the index lives under your home folder.

  1. Install Bun

    The runtime Openings runs on. One command, then close and reopen the window.

    macOS or Linux, in Terminal

    curl -fsSL https://bun.sh/install | bash

    Windows, in PowerShell

    powershell -c "irm bun.sh/install.ps1 | iex"

  2. Install Openings

    In a new Terminal or PowerShell window, so Bun is on your path.

    bun add --global openings

  3. Connect your AI app

    Add this to your app's MCP servers. In Claude Desktop that is Settings → Developer → Edit Config.

    {
  "mcpServers": {
    "openings": { "command": "openings-mcp", "args": [] }
  }
}
Your search, your control.How Openings handles your data and what it can—and can’t—do.Read more +

Read the code

Everything is open source under the MIT licence. The npm package contains exactly the files on GitHub, and you can list them before installing.

Your resume stays in memory

As a connector, your app sends the resume to our server, which reads it for that one request and discards it. It is never written to disk, never kept, and never sent to an employer. Run Openings on your own machine and the text never leaves your computer. How we handle your data.

No recruiters, no fees, no fake jobs

Every listing comes from an employer's own careers site, verified before it enters the index. Every link goes straight there. Nobody in between, and nobody will ever ask you to pay.

It cannot apply for you

There is no form-fill, apply, or submit tool. The only thing Openings writes is your own job index, in a folder on your machine.

What leaves your machine

Requests to public job boards, and a copy of the public job listings you crawl, sent to the shared index so other people's setup is faster. Also an anonymous usage record under a random install ID: which tool ran, the countries, roles, skills and search text you asked for, which jobs you opened, and the skill and title words extracted from your resume. Never the resume itself, your name, or your contact details, and no IP address is kept with it. Set OPENINGS_USAGE=off to stop the usage record, or OPENINGS_AGGREGATOR_URL to an empty string to keep everything on your machine.

No account, no key, no model call

Openings never talks to an AI model itself. Your own agent does the reasoning; Openings gives it evidence.

verify it yourself Jobs by country and title Source code on GitHub Package contents on npm Listing on Smithery Connector health on Glama Quickstart with an example conversation Report a problem Questions: [email protected]