JobSpy MCP Server

Overview

The JobSpy MCP Server bridges the gap between AI assistants and real‑world job markets by exposing a unified, structured interface to multiple job‑listing platforms such as Indeed, LinkedIn, Glassdoor, and more. Instead of having an assistant manually scrape or parse each site, the server runs the open‑source JobSpy tool and returns results in a clean JSON or CSV payload that Claude (or any MCP‑compatible client) can ingest directly. This eliminates repetitive boilerplate code for developers and gives AI agents instant access to up‑to‑date hiring data.

At its core, the server accepts MCP tool invocation requests. A typical request tells the agent to “search for jobs” with filters like , , and a comma‑separated list of target sites. The server then orchestrates the underlying JobSpy script, handles authentication tokens if required, and streams progress back to the client via Server‑Sent Events (SSE). When the search completes, the structured data is returned, ready for downstream processing such as ranking, natural‑language summarization, or feeding into a hiring workflow.

Key capabilities include:

• Multi‑platform coverage: A single call can query dozens of job boards, reducing API fatigue and ensuring broader candidate visibility.
• Fine‑grained filtering: Parameters for search terms, geographic location, posting date, and site selection give developers precise control over the data returned.
• Transport versatility: The server supports both standard input/output (ideal for desktop assistants like Claude) and SSE, enabling real‑time updates in web applications.
• Data normalization: By converting disparate HTML or API responses into a consistent schema, the server removes ambiguity for AI models and downstream analytics.

Real‑world use cases span from personal career coaches who want to surface the latest opportunities for clients, to enterprise recruiters automating candidate sourcing pipelines. A recruiter’s AI assistant can ask the server for “remote software engineer roles in New York posted within the last week,” receive a JSON list, and immediately generate personalized outreach emails. Web portals can subscribe to the SSE endpoint for live job feed widgets, while background services can batch queries on a schedule and trigger alerts when new positions match user profiles.

Because the server is built around the proven JobSpy tool, it inherits robust scraping logic and respects site rate limits. Developers benefit from a single configuration point (environment variables or a file) and the ability to swap Docker images for isolated execution. The result is a lightweight, plug‑and‑play MCP server that empowers AI assistants to act as real‑time job market experts without the overhead of maintaining individual scrapers or APIs.