college-aid-mcp

college-aid-mcp

Searches US colleges and scholarships using government data, enabling comparisons of tuition, debt, earnings, and program-specific outcomes.

Category
访问服务器

README

college-aid-mcp

An MCP server that combines the US College Scorecard (Dept of Education) and CareerOneStop Scholarship Finder (Dept of Labor) so you can ask Claude to research colleges and matching scholarships in one conversation.

What it can do

search_colleges — search colleges by name, state, degree type, or max tuition. Returns for each school:

  • In-state / out-of-state tuition
  • Median debt at graduation
  • Median earnings 10 years after entry
  • 4-year graduation rate
  • Admission rate

search_scholarships — search 9,500+ scholarships and grants. Filter by keyword, state, study level. Returns award amount, deadline, provider, and direct link.

find_similar_colleges — given a reference school, generate a reach/target/safety list of alternatives with similar Carnegie classification. Filters by admission rate windows (reach = more selective, target = ±7%, safety = less selective), then sorts each bucket by earnings, tuition, or closest match. Falls back to adjacent Carnegie tiers if a bucket returns too few results.

compare_colleges — side-by-side comparison table for 2–5 schools. Runs all lookups in parallel and returns a single markdown table across 18 metrics: location, type, acceptance rate, SAT/ACT ranges, in-state/out-of-state tuition, net price at 3 income brackets, median debt, Pell rate, 4-yr grad rate, retention, 10-yr earnings, and first-gen student share.

search_careers — search careers by job title or keyword using CareerOneStop/O*NET data. Returns for each occupation:

  • Annual salary percentiles (P25 / median / P75 / P90), nationally and by state
  • Bright job outlook flag (above-average growth or openings)
  • Typical education required
  • Key job tasks

estimate_loan_repayment — pure computation, no API key needed. Given a loan balance and expected salary, calculates:

  • Standard 10-year plan: monthly payment, total interest
  • Extended 25-year plan: lower monthly cost, higher total interest
  • Income-driven repayment (IDR): payment as % of discretionary income, forgiveness after 20 years if balance remains
  • Debt-to-income health check and a recommendation on which plan makes sense

get_college_details — deep dive on a single school. Returns:

  • Net price broken down by family income bracket ($0–30k, $30–48k, $48–75k, $75–110k, $110k+)
  • SAT/ACT score ranges (25th–75th percentile)
  • Acceptance rate, enrollment, 1st-year retention, 4-year graduation rate
  • First-generation student share, Pell grant rate
  • Median debt at graduation and 10-year earnings
  • Institution type, Carnegie classification, locale

search_by_major — compare the same major across schools using program-specific data, not school-wide averages. Supports 60+ major aliases (e.g. "cs", "nursing", "mechanical engineering") or raw 4-digit CIP codes. Returns for each school:

  • Program-specific 4yr median earnings (vs. national benchmark for that program)
  • Program-specific median student debt
  • In-state tuition
  • Sortable by earnings, debt, or tuition

Example prompts

  • "Find CS-focused universities in Washington state with tuition under $15k and show me STEM scholarships I could apply to as a high schooler."
  • "Compare median debt vs earnings for the top 10 engineering schools."
  • "Find first-generation college student scholarships in California."
  • "Rank nursing bachelor programs in Texas by earnings — which schools beat the national median?"
  • "Show me mechanical engineering programs sorted by lowest debt in the midwest."
  • "Get full details for Georgia Tech — what would I actually pay if my family earns $60k/year?"
  • "Compare Georgia Tech and Carnegie Mellon on net price, SAT ranges, and 10-year earnings."
  • "I'll graduate with $28k in debt and expect to earn $55k — show me my repayment options."
  • "Find CS programs in California sorted by lowest debt, then estimate repayment for the top school."
  • "I'm targeting University of Washington — give me reach, target, and safety alternatives sorted by lowest cost."
  • "Compare MIT, Stanford, Georgia Tech, and Carnegie Mellon side by side on cost, debt, and earnings."
  • "What do software engineers actually earn in Washington state, and which CS programs give the best shot at hitting P75 salary?"
  • "Compare nursing salaries in Texas vs nationally, then find nursing programs with the lowest debt."

Setup

1. Get API keys (both free)

College Scorecard API key

  1. Go to api.data.gov/signup
  2. Register — you'll receive a key by email immediately
  3. Set as env var: COLLEGE_SCORECARD_API_KEY=your_key

CareerOneStop API credentials

  1. Go to api.careeronestop.org/api-explorer
  2. Register for a free account
  3. You'll receive a User ID and Token
  4. Set as env vars: CAREERONESTOP_USER_ID=your_user_id and CAREERONESTOP_TOKEN=your_token

2. Install

git clone https://github.com/Fusingchart/college-aid-mcp.git
cd college-aid-mcp
npm install
npm run build

3. Configure Claude Desktop

Add to your claude_desktop_config.json (usually at ~/Library/Application Support/Claude/claude_desktop_config.json on macOS):

{
  "mcpServers": {
    "college-aid": {
      "command": "node",
      "args": ["/absolute/path/to/college-aid-mcp/build/index.js"],
      "env": {
        "COLLEGE_SCORECARD_API_KEY": "your_scorecard_key",
        "CAREERONESTOP_USER_ID": "your_user_id",
        "CAREERONESTOP_TOKEN": "your_token"
      }
    }
  }
}

4. Configure Claude Code

claude mcp add college-aid \
  -e COLLEGE_SCORECARD_API_KEY=your_key \
  -e CAREERONESTOP_USER_ID=your_user_id \
  -e CAREERONESTOP_TOKEN=your_token \
  -- node /absolute/path/to/college-aid-mcp/build/index.js

Data sources

Source Provider Cost
College Scorecard API US Dept of Education Free
CareerOneStop Scholarship Finder US Dept of Labor Free

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
Kagi MCP Server

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。

官方
精选
Python
graphlit-mcp-server

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。

官方
精选
TypeScript
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选
mcp-server-qdrant

mcp-server-qdrant

这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。

官方
精选