Packages

26403 results

UI-TARS

Automation

UI-TARS is an open-source multimodal agent for automated GUI interaction. It uses a vision-language model to parse/ground visual observations and generate structured action instructions that can be translated into automation code (e.g., PyAutoGUI) to operate desktop/mobile UIs.

Skill Priority Stale ai-ml automation computer-use
42

agent-skills-hub

DevTools

Agent Skills Hub appears to be a web app and API that periodically syncs GitHub repositories, cleans/deduplicates and classifies them, computes multi-dimensional quality scores (including an “Agent Readiness” dimension), stores results in Supabase, and serves listings, trending/rising/top-rated views, RSS/sitemap, plus endpoints for submitting skills and subscribing to a newsletter.

Priority Stale ai-ml devtools api
42

apache-hadoop-hive-mcp-server-by-cdata

Ai Ml

Provides a local, read-only Model Context Protocol (MCP) server that exposes Apache Hadoop Hive data via CData JDBC Driver as MCP tools (get tables, get columns, run SELECT queries), so local LLM clients (e.g., Claude Desktop) can query live Hive data without writing SQL.

MCP Stale mcp stdio data-connector
42

claude-cracks-the-whip

DevTools

A Claude Code “skill” that orchestrates multiple AI coding agents to execute parallel development tasks, using tmux panes, background processes, and/or MCP. It deploys assignments, collects full work logs, inspects diffs and quality checks, and can send correction assignments when results are sloppy.

Skill Priority Stale ai-agents ai-orchestration claude-code-skill
42

evals

Ai Ml

OpenAI Evals is an open-source framework for evaluating LLMs and LLM systems. It provides an existing registry of benchmark/evaluation definitions (often data-driven), tooling to run local evaluations, and guidance for creating custom evals (including model-graded evals via YAML/templates).

Skill Priority Stale ai-ml evaluation llm-evals
42

llm-workflow-engine

Ai Ml

LLM Workflow Engine (LWE) is a Python-based CLI and workflow manager for building and running LLM interactions (chat/tool use) from the shell, with a plugin architecture and support for multiple LLM providers (including OpenAI via the ChatGPT API).

Skill Priority Stale ai-ml llm cli
42

mcp-servers

Security

Monorepo providing multiple TypeScript Model Context Protocol (MCP) server packages for Check Point security platforms (e.g., management, management logs, threat prevention, HTTPS inspection, Harmony SASE, reputation, gateway CLI diagnostics, threat emulation, GAIA, documentation assistant, CPInfo analysis, Argos ERM, and policy insights). Each MCP server is intended to expose a structured interface to query and analyze security-related data.

MCP Priority Stale mcp check-point security
42

MCPtrace

Infrastructure

Provides an MCP (Model Context Protocol) server that generates and executes bpftrace kernel tracing programs from AI assistant requests, including discovery/listing of available probes and asynchronous execution with later polling for results.

MCP Priority Stale mcp ebpf bpftrace
42

mcu-mcp-server

Ai Ml

An MCP (Model Context Protocol) server implemented with Express.js that exposes a tool to generate a color scheme from a provided seed color using Google’s Material Color Utilities library.

MCP Stale mcp mcp-server express
42

oracle.oci-fusion-applications-mcp-server

Cloud

MCP server for Oracle OCI Fusion Applications Environment Management, enabling AI agents to manage Oracle Fusion Applications environments, schedule maintenance, manage upgrades, and control Fusion Applications lifecycle within OCI.

MCP Free tier Stale oracle oci fusion-applications
42

prowlarr

Automation

Prowlarr is an application that manages and indexes torrent downloads by integrating with multiple torrent indexers and mapping them to download clients. It automates searching, prioritization, and synchronization of indexer categories and releases.

Priority Stale self-hosted automation torrent-indexer-management
42

semantic-router

Ai Ml

Semantic Router (vLLM Semantic Router) provides a system-level routing layer for mixture-of-models by classifying or scoring incoming requests/contexts and dispatching them to appropriate models or pipelines.

Skill Priority Stale ai-ml llmrouter mixture-of-models
42

sufy-mcp-server

DevTools

An MCP (Model Context Protocol) server package named sufy-mcp-server, intended to expose “Sufy” capabilities as MCP tools so AI agents can call them in a standardized way.

MCP Stale mcp ai-agents server
42

tc-weather-mcp-server

Infrastructure

tc-weather-mcp-server is an MCP server intended to provide weather-related capabilities (typically backed by an external weather provider) to AI agents via the Model Context Protocol.

MCP Stale mcp weather tooling
42

zed-mui-mcp-server

DevTools

Provides a Zed extension that exposes an MCP server for retrieving Material UI (MUI) documentation data inside Zed.

MCP Stale zed mcp material-ui
42

zopen-mcp-server

Infrastructure

A Go-based Model Context Protocol (MCP) server that exposes a set of `zopen` command-line operations (list/query/install/remove/upgrade/info/version) as MCP tools. It supports stdio-based local execution (run tools on the same machine) and an optional remote mode that uses SSH to run `zopen` against a target z/OS system.

MCP Stale mcp zopen zos
42

cloudwatch-mcp-server

Monitoring

Provides an MCP server that exposes AWS CloudWatch functionality (metrics/logs/alarms, depending on supported tools) to AI agents via MCP.

MCP Stale mcp aws cloudwatch
42

Flow-Like

Automation

A strongly-typed, local-first visual workflow automation platform built in Rust. Provides a drag-and-drop editor with 900+ built-in nodes for APIs, databases, file processing, and AI models. Runs on laptops, phones, servers, or edge devices with full offline capability and claims ~244K workflows/second throughput.

Stale workflow automation rust
42

jebmcp

Ai Ml

jebmcp provides an MCP server/plugin that integrates the JEB Pro reverse-engineering environment with MCP-capable clients (e.g., Cline/Cursor/RooCode). It starts a local HTTP endpoint (127.0.0.1:16161) and exposes MCP methods for querying/manipulating decompiled/exported artifacts in JEB, including class/method/field inspection and renaming.

MCP Priority Stale mcp mcp-server jython
42

lnd

Payments

lnd is an implementation of the Lightning Network node. It runs as a daemon that connects to the Bitcoin network, manages channels with peers, and provides an RPC/REST-like interface (via gRPC/proxy servers) for creating/monitoring payments over the Lightning Network.

Stale ai-ml infrastructure payments
42
8642
Packages Evaluated
17761
Need Evaluation
9228
Need Re-evaluation
Community Powered