# Opportunity MCP [Health: Active]

**Category:** 🎓 Education  
**Repository:** https://github.com/revolutionarybukhari/opportunity-mcp  
**GitHub Stars:** 1  
**Views:** 2  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/opportunity-mcp

## Description
Search youth opportunities (scholarships, fellowships, internships) from any AI assistant.

## Tools
Capabilities this server exposes over MCP:

- **search_opportunities** — Full-text search across all indexed opportunities with optional filters.
- **get_opportunity** — Retrieve full details for a single opportunity by its ID.
- **list_latest** — Newest opportunities across all sources, sorted by post date.
- **list_upcoming_deadlines** — Opportunities closing within `N` days, sorted by deadline.
- **list_sources** — List indexed sources, item counts, and last-refresh timestamps.
- **refresh_index** — Re-fetch sources on demand. Optional `source` argument limits the refresh to one site.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `uvx` (confidence: high):

```json
"mcpServers": {
  "opportunity-mcp": {
    "command": "uvx",
    "args": ["opportunity-mcp"]
  }
}
```

## Documentation

## What the Opportunity MCP MCP server does

Opportunity MCP MCP server turns a local index of publicly available youth opportunities into MCP tools for an AI assistant. The indexed content includes scholarships, fellowships, internships, conferences, exchange programs, competitions, grants, awards, and other listings. Records are collected from opportunity-discovery sites and link back to their original articles.

The current live RSS sources include Opportunities Corners, Opportunities for Youth, Opportunity Desk, Scholarships Corner, Opportunities Circle, Opportunities for Africans, and Scholars4Dev. Youth Opportunities and After School Africa are listed as planned HTML adapters rather than active sources.

## How it works

The server runs locally and communicates over stdio. A FastMCP layer receives tool calls and queries a SQLite database using FTS5 for text search. Separate source adapters fetch and normalize listings into validated opportunity records, while the query layer works against the normalized index rather than individual websites.

The index is refreshed through a scheduled process at most every six hours, and users can also request a refresh through the server. Queries are processed on the user's machine; the only external requests described by the project are periodic fetches from source sites. Source adapters respect robots.txt, identify themselves with a project URL, and cap summaries at 500 characters.

## Setup and configuration

Install the package from PyPI, then build the local index once with `opportunity-mcp-refresh`. Configure an MCP client to launch `opportunity-mcp` as its command. Clients that support `uvx` can instead run `uvx opportunity-mcp` without a separate local installation step.

For development, clone the repository, run `uv sync` or install the editable development package, build the index with `uv run opportunity-mcp-refresh`, and start the server with `uv run opportunity-mcp`. The documented transport is local stdio; a hosted Streamable HTTP endpoint is planned, not described as currently available.

## Tools and capabilities

The Opportunity MCP MCP server exposes six JSON-based tools with Pydantic-typed results:

- `search_opportunities` performs full-text searches with optional opportunity type, funding, deadline, and result-limit filters.
- `get_opportunity` returns the complete record for a supplied opportunity ID.
- `list_latest` returns recently posted opportunities, optionally filtered by type.
- `list_upcoming_deadlines` finds listings closing within a specified number of days and sorts them by deadline.
- `list_sources` reports indexed sources, item counts, and their latest refresh timestamps.
- `refresh_index` fetches sources again, optionally limiting the operation to one source.

Supported types are scholarship, fellowship, internship, conference, exchange, competition, grant, award, and other. Country, education level, and language filtering are not currently built into the server; the roadmap says those fields are currently delegated to the AI client.

## Limitations and notes

This is alpha software at version 0.1.x. The schema, tool surface, and source list can change as adapters develop, so the README recommends pinning a minor version in production. Source coverage is dependent on the listed sites and their feeds. In particular, the Scholars4Dev adapter is live but its upstream feed is currently empty, while the two HTML sources remain planned.

The server is designed for local use and does not provide a hosted remote endpoint yet. It does not offer saved profiles or weekly digests; those are roadmap items. Indexed data is public, and source removals requested by site owners are intended to be honored within 24 hours.

_Full upstream README: https://allmcps.com/mcp/opportunity-mcp/readme_