indeed-mcp

# Indeed MCP Server

<!-- mcp-name: com.hasdata/indeed -->

A hosted Model Context Protocol (MCP) server that gives Claude, Cursor, Windsurf and any other MCP client two read-only Indeed tools. Search job listings by keyword and location, and read a single posting in full, all as structured JSON, with no Indeed developer account and no partner approval.

It reads public job postings that a signed-out visitor can see.

```
https://mcp.hasdata.com/api/mcp?apis=indeed
```

[![Glama score](https://glama.ai/mcp/servers/HasData/indeed-mcp/badges/score.svg)](https://glama.ai/mcp/servers/HasData/indeed-mcp)
[![tool contract](https://github.com/HasData/indeed-mcp/actions/workflows/contract.yml/badge.svg)](https://github.com/HasData/indeed-mcp/actions/workflows/contract.yml)
[![MCP](https://img.shields.io/badge/MCP-remote%20%7C%20streamable%20HTTP-6366f1?style=flat-square)](https://modelcontextprotocol.io)
[![Tools](https://img.shields.io/badge/tools-2-10b981?style=flat-square)](#tools)
[![License](https://img.shields.io/badge/license-MIT-blue?style=flat-square)](LICENSE)

## Contents

- [What you need](#what-you-need)
- [Quick start](#quick-start)
- [Example prompts](#example-prompts)
- [Tools](#tools)
- [Errors and failure paths](#errors-and-failure-paths)
- [Pricing, free tier and limits](#pricing-free-tier-and-limits)
- [Tool selection](#tool-selection)
- [How it compares](#how-it-compares)
- [FAQ](#faq)
- [HasData links](#hasdata-links)
- [Development](#development)
- [Contributing](#contributing)
- [License](#license)

## What you need

An MCP client and a HasData API key from the [dashboard](https://app.hasdata.com/sign-up?utm_source=github&utm_medium=syndication&utm_campaign=indeed-mcp), free to create with no card, and the trial covers about 200 calls at the 5-credit rate. This is a remote server, so the simplest path is a URL and an `x-api-key` header, with no container to run and no Indeed developer account anywhere in the flow. A client that only speaks stdio reaches it through a thin launcher, published as `@hasdata/indeed-mcp` on npm and `hasdata-indeed-mcp` on PyPI, shown below.

## Quick start

The server URL is the same for every client. We run it hands-on in Claude Code and Claude Desktop. The other blocks follow each client's own documented format for a remote server.

| Field | Value |
| :--- | :--- |
| URL | `https://mcp.hasdata.com/api/mcp?apis=indeed` |
| Transport | HTTP, streamable |
| Auth header | `x-api-key: HASDATA_API_KEY` |

Clients with OAuth support can add the same URL as a connector and sign in without putting a key in a config file.

<details>
<summary><b>Claude Code</b></summary>

```bash
claude mcp add --transport http indeed "https://mcp.hasdata.com/api/mcp?apis=indeed" \
  --header "x-api-key: HASDATA_API_KEY"
```

</details>

<details>
<summary><b>Claude Desktop</b></summary>

Settings, then Connectors, then Add custom connector, then paste `https://mcp.hasdata.com/api/mcp?apis=indeed` and sign in.

For the config-file route, Claude Desktop loads only local (stdio) servers, so it reaches a remote server through a stdio launcher. The `@hasdata/indeed-mcp` package is that launcher, and it reads the key from the environment. Add this to `claude_desktop_config.json`:

```json
{
  "mcpServers": {
    "indeed": {
      "command": "npx",
      "args": ["-y", "@hasdata/indeed-mcp"],
      "env": { "HASDATA_API_KEY": "YOUR_KEY" }
    }
  }
}
```

For Python instead of Node, swap the launcher for the PyPI package, which `uvx` runs without a manual install:

```json
{
  "mcpServers": {
    "indeed": {
      "command": "uvx",
      "args": ["hasdata-indeed-mcp"],
      "env": { "HASDATA_API_KEY": "YOUR_KEY" }
    }
  }
}
```

</details>

<details>
<summary><b>Cursor</b></summary>

`~/.cursor/mcp.json` for every project, or `.cursor/mcp.json` for one:

```json
{
  "mcpServers": {
    "indeed": {
      "url": "https://mcp.hasdata.com/api/mcp?apis=indeed",
      "headers": { "x-api-key": "HASDATA_API_KEY" }
    }
  }
}
```

</details>

<details>
<summary><b>Windsurf</b></summary>

`~/.codeium/windsurf/mcp_config.json`. Windsurf calls the field `serverUrl`, not `url`:

```json
{
  "mcpServers": {
    "indeed": {
      "serverUrl": "https://mcp.hasdata.com/api/mcp?apis=indeed",
      "headers": { "x-api-key": "HASDATA_API_KEY" }
    }
  }
}
```

</details>

<details>
<summary><b>VS Code</b></summary>

`.vscode/mcp.json` in the workspace:

```json
{
  "servers": {
    "indeed": {
      "type": "http",
      "url": "https://mcp.hasdata.com/api/mcp?apis=indeed",
      "headers": { "x-api-key": "HASDATA_API_KEY" }
    }
  }
}
```

</details>

## Example prompts

Prompts, not code. Paste one in and the agent picks the tool itself. Each is annotated with the calls it takes, because every successful call costs 5 credits.

> Search Indeed for "python developer" jobs in New York sorted by date, and give me the ten most recent with company and salary.

*One call, 5 credits. A listing page carries company, salary and posted date already.*

> Take the top posting from that search and pull its full description and requirements.

*One call, 5 credits. The listing carries a job URL, which the details tool takes directly.*

> Find "data analyst" jobs in Austin, then pull full details on the three that list a salary.

*Four calls, 20 credits. One listing, then one details call for each of the three.*

> Compare the salaries posted for "registered nurse" in Chicago against Houston.

*Two calls, 10 credits, one listing per city.*

Salary is on a listing only when the posting states one, so a "jobs with salary" prompt filters on the field rather than assuming it. Paging costs a call each time, through the `start` offset.

## Tools

Two tools, read-only. Samples below are trimmed from real calls, and the numbers move as Indeed updates. Read them as shapes. Each tool name links to its endpoint reference, which carries the full field list.

The samples are the payload, not the whole response. A `tools/call` result carries one text block, and that text is itself JSON holding `url`, `status`, `text` and `json`, with the scraped data under `json`. From a raw JSON-RPC response the path is `result.content[0].text`, parsed, then `.json`. A chat client unwraps that for you and code talking to the endpoint directly does not.

### Get Indeed job listings

[`hasdata_indeed_listing_getJobListings`](https://docs.hasdata.com/apis/indeed/listing?utm_source=github&utm_medium=syndication&utm_campaign=indeed-mcp)

A page of search results by keyword and location.

| Parameter | Type | Required | Notes |
| :--- | :--- | :--- | :--- |
| `keyword` | string | yes | The search term, as a candidate would type it |
| `location` | string | yes | City and state, or any location string Indeed accepts |
| `sort` | string | | `relevance` by default, or `date` for newest first |
| `domain` | string | | A country site such as `au.indeed.com`. Defaults to the US site |
| `start` | number | | Result offset for the next page, a number and not a URL. Read the offset from `pagination.nextPage` in the previous response |

Returns `searchInformation`, a `jobs` array, `peopleAlsoSearchFor`, and `pagination` whose `nextPage` is the URL of the following page. Each job carries `title`, `company`, `location`, `url`, a short `description`, `sponsored`, the relative `date` and the absolute `isoDate`, a `details` array of labels like `Full-time` and `Hybrid work`, a `benefits` array, and a `salary` object when the posting states one.

> A listing carries `details` as an array of plain strings. The job-details tool below returns `details` as an object with `jobType` and `workSetting` arrays. Same name, different shape, so read each per its own tool.

```json
{
  "title": "Hedge Fund Application Developer",
  "company": "TBA",
  "location": "Stamford, CT 06901",
  "url": "https://www.indeed.com/pagead/clk?...",
  "sponsored": true,
  "date": "30+ days ago",
  "isoDate": "2025-06-03T17:22:54.869Z",
  "details": ["Full-time", "Hybrid work"],
  "benefits": ["Health insurance", "401(k)", "401(k) matching"],
  "salary": { "min": 80000, "max": 150000, "type": "YEARLY" }
}
```

### Get Indeed job details

[`hasdata_indeed_job_getJobDetails`](https://docs.hasdata.com/apis/indeed/job?utm_source=github&utm_medium=syndication&utm_campaign=indeed-mcp)

One posting in full, by its job URL.

| Parameter | Type | Required | Notes |
| :--- | :--- | :--- | :--- |
| `url` | string | yes | A posting URL, the `url` field from a listing result |

Returns `title`, `company`, `location`, `sponsored`, a `details` object with `jobType` and `workSetting` arrays, a `salary` object when present, and the posting body in both `description` (plain text) and `descriptionHtml` (the same content with its list and paragraph markup kept). Use `descriptionHtml` when you want the structure and `description` when you want to feed clean text to a model.

```json
{
  "title": "Python Developer",
  "company": "Think IT Technologies",
  "location": "New York, NY 10114",
  "sponsored": false,
  "details": { "jobType": ["Contract"], "workSetting": ["In-person"] },
  "salary": { "min": 60.5, "max": 65, "type": "HOURLY" },
  "description": "Overview\nWe are seeking a Python Developer...",
  "descriptionHtml": "<p><b>Overview</b></p><p>We are seeking a Python Developer...</p>"
}
```

## Errors and failure paths

Your client almost never sees an HTTP error code from a tool call. The MCP layer answers 200 and puts the failure inside the result, with `isError` set to `true` and the reason as text. The agent reads a message where you might expect a status line.

**A wrong key surfaces as tool output, not as a failed connection.** `tools/list` accepts any non-empty key and returns both tools, so the client completes its handshake and shows green. The first tool call then comes back with `isError: true` and the text `HasData API error: 401 Unauthorized`. Watch for that string, because nothing earlier in the flow repor