Packages

An experimental MCP server (Rust) that exposes a cat “breeds” search route (and currently only that route per the README).

mlflow-server is the server component of MLflow, providing an HTTP API to manage machine learning experiments, runs, artifacts, models, and the tracking/registry services.

OpenCompass is an open-source LLM evaluation platform. It provides configurable evaluation pipelines (via CLI and Python scripts) to run model benchmarks across many datasets, including support for local/open-source models and API-based models (e.g., OpenAI/Qwen), with optional inference acceleration backends (e.g., vLLM, LMDeploy).

peerflix-server is a server component for peerflix, enabling streaming video via WebTorrent/P2P from a server process while exposing a local HTTP/web interface for playback.

An MCP (Model Context Protocol) server wrapper intended to expose “shuidi” (shuidi-data) data access to AI agents via MCP tools. The provided prompt does not include README/repo contents, so the exact tools, endpoints, parameters, and operational behavior cannot be verified from the given information.

SkyWalking OAP Server (Observability Analysis Platform) ingests, processes, and stores distributed tracing/metrics/logging data for Apache SkyWalking. It serves as the backend that receives telemetry from agents/collectors, performs analysis/aggregation, and provides data access for the SkyWalking UI/other consumers.

tcpping-mcp-server is an MCP server intended to perform TCP reachability checks (“ping” over TCP) to specified hosts/ports, exposing the functionality to AI agents via the Model Context Protocol.

weather-mcp-server-wl appears to provide an MCP (Model Context Protocol) server for retrieving weather information for use by AI agents.

Provides a CLI tool to wrap an existing API or MCP server with x402 payment verification, analytics, and rate limiting, then deploy the resulting monetized gateway to supported platforms (e.g., Vercel/Railway/Fly.io/Docker).

ybl-mcp-server is an MCP server package intended to expose tools/resources to AI agents via the Model Context Protocol, likely backed by YBL-related capabilities. From the provided input, no README/repo manifest details are available, so tool coverage, parameters, and operational behavior cannot be verified.

MCP server that provides Gaode (AMap) weather functionality to agents via Model Context Protocol tools (exact weather endpoints/fields are not provided in the supplied prompt).

zipkin-server is an open-source Zipkin backend service that receives, stores (optionally), and serves distributed tracing data (spans) collected from instrumented applications via Zipkin-compatible APIs.

EmulatorJS is a JavaScript-based framework for running software emulated in the browser (commonly via WebAssembly), providing an in-browser execution environment for classic systems/ROMs and related tooling.

Provides an IPv6-focused hosts file (and scripts) intended to speed up access by mapping domains (e.g., Google/YouTube/Facebook/Wikipedia) to their corresponding IPv6 addresses. Includes a Python script (update_hosts.py) to query DNS (default AAAA) and regenerate a hosts file, and a shell script (merge_snippets.sh) to combine host snippets.

mcp-oauth-gateway is an experimental OAuth 2.1 authorization layer for MCP servers. It fronts MCP endpoints with an OAuth authorization server (GitHub as IdP) and supports dynamic client registration, user login (web/device flows), token issuance (JWT/refresh), and forwarding/authentication for MCP HTTP transport, without modifying the upstream MCP server code (wrapping/bridging for HTTP transport is used).

Provides an MCP server (implemented in Java/Spring) that uses AI to generate articles and automates publishing them to CSDN, Juejin, and Cnblogs via MCP tool functions (search for content and publish articles).

qBittorrent (qbittorrent) is a self-hostable BitTorrent client providing a web UI and optional remote control interface to manage torrents (add/remove, start/stop, view status, etc.).

q-mem-mcp-server is an MCP server package intended to expose “memory” (likely a Q-* memory component) to an MCP-capable agent runtime via the Model Context Protocol.

qwen-browser-use-mcp-server is an MCP (Model Context Protocol) server that exposes Qwen/Browse-automation browser-use capabilities to MCP clients (agents). It is intended to let an AI agent interact with a browser environment through MCP tools, presumably using Qwen as the underlying model for reasoning and browse assistance.

simctl-mcp-server is an MCP (Model Context Protocol) server that exposes functionality related to iOS Simulator tooling (simctl) to agent/LLM runtimes via MCP tools. It is intended to let an agent inspect/control simulator state as part of automated workflows.