Algora MCP Server

What the idapixl/algora-mcp-server MCP server does

The idapixl/algora-mcp-server MCP server connects AI agents to Algora's public bounty platform. It is intended for discovering open-source work that has an associated reward, checking an organization's available bounties, and producing simple market summaries.

The server supports five operations. Agents can browse bounties with filters, retrieve bounties belonging to a specific organization, search titles and descriptions by keyword, rank opportunities by reward, and request aggregate statistics. This makes it suitable for questions such as finding TypeScript tasks above a chosen dollar amount or comparing bounty activity across organizations and technologies.

How it works

Communication uses the standard MCP stdio transport. The local process reads requests from the MCP client's standard input and writes responses to standard output; it does not start an HTTP server or listen on a port.

API requests are handled through Algora's public API using the @algora/sdk tRPC client. The README also states that keyword search and minimum or maximum amount filtering can be performed in memory when those operations are not provided directly by the API. No authentication configuration is required for the public data access.

Setup and configuration

The idapixl/algora-mcp-server MCP server requires Node.js 18 or later. The documented local setup is to clone the repository, install its npm dependencies, and build the TypeScript source with npm run build. The resulting entry point is dist/index.js.

MCP clients are configured to launch that file with Node. Claude Desktop, Cursor, and Windsurf examples all use a node command with the built file as its argument; the example requires replacing the repository path with the actual absolute path on the host. Development can instead use npm run dev, while npm run typecheck checks types without emitting build output.

Tools and capabilities

The five exposed tools are:

• list_bounties: Lists bounties with status, organization, reward range, technology, result limit, and pagination-cursor options. The default status is open, and the documented limit range is 1 to 100.
• get_org_bounties: Retrieves bounties for a required organization slug, with optional status and result limit settings.
• search_bounties: Searches across bounty titles, descriptions, and repositories using a required keyword. Technology, minimum reward, and result limit filters are also available.
• get_top_bounties: Returns higher-value bounties ordered by reward, with optional technology, minimum amount, and limit parameters. Its documented default minimum is 100 USD.
• get_bounty_stats: Returns aggregate information including total value, counts, and breakdowns by technology and organization. An organization can be supplied to scope the statistics.

Limitations and notes

The idapixl/algora-mcp-server MCP server is focused on reading and analyzing bounty data. The provided material documents discovery, filtering, ranking, searching, and aggregation tools, but does not describe tools for claiming bounties, submitting work, or modifying Algora records.

It depends on Algora's public API and requires network access to retrieve current results. No API key, port configuration, or environment variable is documented. Because it runs locally over stdio, each MCP client must be configured with the path to the built JavaScript entry point.