job-search-mcp
MCP server that exposes job search data from multiple boards, enabling clients to query and manage job listings via natural language.
README
Job Search Tool
Local-first job search automation: collect jobs from multiple boards, score them against your own criteria, store them in SQLite, review them in a React dashboard, query them through REST, and expose the same data to MCP clients.
Features
- Multi-board ingestion through JobSpy.
- Configurable keyword scoring with separate save and notification thresholds.
- SQLite persistence with bookmarks, applied state, blacklist, and retention.
- React dashboard for review, filtering, analytics, cleanup, and CSV export.
- Local semantic search through ChromaDB's bundled ONNX embedder.
- Telegram notifications for high-scoring new jobs.
- Unified ASGI web server for dashboard, REST API, and MCP.
- Docker Compose deployment with localhost-first network defaults.
Docker Quick Start
-
Create your runtime configuration:
cp config/settings.example.yaml settings.yaml -
Edit
settings.yaml, especiallylocations,queries,scoring, and optional Telegram settings. -
Start the scheduler and web server:
docker compose up -d -
Open the dashboard:
http://127.0.0.1:8501
The same web process exposes REST at http://127.0.0.1:8501/api and MCP at
http://127.0.0.1:8501/mcp. Published ports bind to 127.0.0.1 by default.
For trusted LAN use, set JOB_SEARCH_WEB_BIND=0.0.0.0 in .env and keep
firewall rules tight.
Interfaces
| Surface | Command | Default URL | Notes |
|---|---|---|---|
| Scheduler | job-search scheduler |
none | continuous search loop |
| Web dashboard | job-search-web |
http://127.0.0.1:8501 |
human review UI |
| REST API | job-search-web |
http://127.0.0.1:8501/api |
optional JOB_SEARCH_API_TOKEN |
| MCP | job-search-web |
http://127.0.0.1:8501/mcp |
streamable HTTP |
Documentation
Local Development
uv sync
npm --prefix frontend install
cp config/settings.example.yaml config/settings.yaml
uv run pytest
npm --prefix frontend run test
uv run pre-commit run --all-files
Local commands run through installed package entrypoints:
uv run job-search once
uv run job-search scheduler
uv run job-search-web
Releases
Docker images are published to Docker Hub by the release workflow when a v*
tag is pushed. Release tags produce semver tags, latest, SBOM, and provenance
metadata for vincenzoimp/job-search-tool.
License
MIT. See LICENSE.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。