Remote Jobs MCP Server
Official# Jobicy Remote Jobs MCP Server
Official MCP server for [Jobicy](https://jobicy.com), providing AI assistants with access to thousands of remote jobs from leading companies worldwide across 20+ professional industries.
AI clients such as Claude, Cursor, Cline, and other MCP-compatible tools can search, filter, and retrieve real-time remote jobs directly from the Jobicy database.
## Available Tools
- `get_jobs`: Fetches a structured list of remote jobs. Supports filtering by count, region (`geo`), category (`industry`), and keywords (`tag`).
- `get_taxonomies`: Retrieves valid filter slugs for regions (`locations`) or categories (`industries`).
---
## Usage
You can connect to this server in three ways: via our hosted public endpoint, via legacy SSE, or by running it locally using `stdio`.
### 1. Streamable HTTP (Recommended)
The default transport for all modern MCP clients (Claude.ai, Cursor, Cline, etc.).
Add this to your `claude_desktop_config.json` or equivalent:
```json
{
"mcpServers": {
"jobicy-jobs": {
"url": "https://jobicy.com/mcp"
}
}
}
```
### 2. Legacy SSE Endpoint
For older MCP clients that require SSE transport:
```json
{
"mcpServers": {
"jobicy-jobs": {
"url": "https://jobicy.com/mcp/sse",
"type": "sse"
}
}
}
```
### 3. Local Installation (via Stdio)
If you prefer to run the server locally on your machine:
1. Clone this repository.
2. Install dependencies: `npm install`
3. Add the server to your configuration using `node` and the `--stdio` flag:
```json
{
"mcpServers": {
"jobicy-jobs-local": {
"command": "node",
"args": ["/path/to/your/cloned/folder/server.js", "--stdio"]
}
}
}
```
### License
MIT
---
[](https://glama.ai/mcp/servers/Jobicy/remote-jobs-mcp-server)
TDQS
Scored across 2 tools
get_jobs retrieves job listings while get_taxonomies retrieves filter metadata, so the two tools have clearly distinct purposes. There is no overlap or ambiguity between them.
Both tools follow the same get_ noun pattern with snake_case naming. The names are predictable and clearly indicate that these are read-only retrieval operations.
Two tools is minimal, but it is not unreasonable for a narrow remote job lookup server. The count feels slightly thin rather than bloated or excessive.
The server covers the core read-only workflow: discovering valid taxonomy slugs and fetching paginated job listings with full descriptions. There could be additional conveniences like job detail lookup, but the primary use case is fully supported.