# OctoCounts MCP Server [Health: Active]

**Category:** 🔄 Version Control  
**Repository:** https://github.com/huanglizhuo/OctoCounts  
**GitHub Stars:** 19  
**npm Downloads (last month):** 7  
**Views:** 1  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/octocounts-mcp-server

## Description
Analyze and compare public GitHub repo line counts (SLOC) via OctoCounts, from any MCP client.

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

```json
"mcpServers": {
  "octocounts-mcp-server": {
    "command": "npx",
    "args": ["-y","octocounts"]
  }
}
```

## Documentation

## What OctoCounts MCP Server MCP server does

OctoCounts MCP Server MCP server makes OctoCounts repository-size data available to agents and developer assistants through MCP. The underlying service analyzes public GitHub repositories and reports source lines of code (SLOC) by language. Results can include the number of files, total lines, code lines, comment lines, blank lines, and languages found.

This is useful when an agent needs a measurable view of repository size rather than GitHub's language-percentage display. It can support analysis and comparison of public repositories, including checking whether a codebase is relatively small, large, or dominated by a particular language.

## How it works

The service resolves a repository reference such as a branch, tag, or commit SHA. It downloads the GitHub archive for that revision instead of cloning the repository, avoiding Git history and reducing the amount of data needed for counting. The archive is then processed with tokei, which produces per-language counts for files, code, comments, blanks, and total lines.

Reports are associated with the repository owner, repository name, resolved commit, and tokei version. This lets repeated requests for the same revision use cached data rather than starting a new count. Analysis jobs are queued so concurrent requests are handled through a job system instead of all running immediately.

## Tools and capabilities

OctoCounts MCP Server MCP server is intended for clients that need to retrieve or use OctoCounts SLOC reports.

- Analyze a public GitHub repository's SLOC.
- Compare repository line-count results.
- Inspect counts broken down by programming language.
- Retrieve file, total-line, code, comment, blank, and language totals.
- Work against a specific branch, tag, or commit reference.
- Reuse cached results for an already analyzed commit.

The source project also exposes reports through a web app, browser extension, CLI, GitHub Action, badges, and HTTP API. Those surfaces are part of OctoCounts generally and should not be assumed to be MCP tools unless the MCP documentation confirms the mapping.

## Limitations and notes

The documented analysis scope is public GitHub repositories. It also does not provide installation commands, transport details, configuration fields, environment variables, or named MCP client compatibility information.

Counts describe what tokei reports for the selected archive and revision. Results may differ when the branch, tag, or commit changes. Because the cache is tied to a resolved commit and tokei version, a new revision or counting-version change can produce a separate report. Use a pinned commit when reproducibility matters.

The repository is licensed under MIT. The README identifies Rust, Axum, Tokio, SQLx, Postgres, and tokei in the backend stack, but those implementation details do not by themselves define requirements for an MCP client.

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