Packages

Toolbase is a macOS desktop application that helps users discover, install, and manage tools/plugins for Claude and other AI platforms using the Model Context Protocol (MCP). It consists of an Electron desktop app (UI and tool management) and a local runner (built with Deno) that executes MCP servers locally to provide tools to Claude Desktop.

Velocity is a local-first web app (Next.js + Bun) that acts as a command center for AI coding CLIs (Claude Code, Codex CLI, Gemini CLI). It runs and inspects coding sessions, captures analytics (usage/cost/latency/tools/routing context), and helps turn repeated patterns into reusable agents/workflows/skills/commands, including routing visualization. It also supports MCP (via the @modelcontextprotocol/sdk dependency) as part of its build layer surfaces.

MCP server package ("wellarchitected-mcp-server") intended to expose capabilities to LLM agents via the Model Context Protocol. It likely provides tools related to AWS Well-Architected guidance, but no repository contents were provided here to confirm exact tools or behavior.

Developer Kit for Claude Code: a modular marketplace of plugins (core + language/infra/tooling plugins) that provide reusable skills, agents, and slash commands to drive a repeatable development workflow (brainstorm/spec-to-tasks/task management/feature development/code review & debugging).

Grapheteria is a Python framework for building and orchestrating agent workflows with a structured model, offering both a code-style API (nodes and edges) and a UI that stays in sync for visual design/debugging, plus features like state tracking, logging, time-travel debugging, and production-oriented execution/scale concepts.

mcp-server-figma is an MCP (Model Context Protocol) server that connects an AI agent to Figma so the agent can fetch or work with design resources via MCP tools.

mcp-server-slack is an MCP (Model Context Protocol) server that integrates a client/agent with Slack, enabling the agent to perform Slack-related actions and/or fetch information via MCP tools.

Neuralangelo is an official implementation for high-fidelity neural surface reconstruction, built on the NVIDIA Imaginaire library. It provides training and inference (including isosurface/mesh extraction) for reconstructing a 3D surface from images/video with known camera poses.

Provides an MCP server (JSON-RPC over HTTP) that exposes methods (e.g., codex_completion, write_code, explain_code, debug_code) to route requests to the OpenAI Codex CLI for code generation, explanation, and debugging. Designed for use with tools like Claude Code.

orange-mcp-server-snowflake is an MCP server intended to let an agent connect to Snowflake and perform common Snowflake operations through MCP tools (e.g., issuing SQL/queries and returning results).

Provides an MCP (Model Context Protocol) server interface for working with PostgreSQL—allowing an AI agent to connect to a Postgres instance and perform database-related operations via MCP tools.

A repository-architecture focused MCP server that exposes tools to an MCP-capable agent, enabling the agent to inspect and reason about a codebase/repo layout/structure for planning or guidance.

Swarms Tools is a Python package that provides a collection of pre-built utilities/tools intended for agent use. Based on the README structure and examples, it includes finance/market data tools (e.g., HTX, stock news, Yahoo Finance, CoinGecko, DeFi TVL), social/media integrations (Telegram, Twitter/X, web scraping), and developer tools (GitHub automation, a code executor), plus a tool orchestration helper (tool_chainer) for sequential/parallel execution.

MCP server for TON-related functionality, exposing TON capabilities as MCP tools to AI agents.

An MCP server that provides WHOIS lookup capabilities to AI agents (e.g., to query domain registration/registrar and related WHOIS records) via MCP tools.

A Go-based Windows screenshot server that captures window/desktop regions (by title/class/pid/handle) and provides both REST endpoints and MCP (JSON-RPC 2.0) methods for screenshot and window/chrome discovery/capture. It also offers real-time WebSocket streaming and Chrome DevTools-based tab capture (requires launching Chrome with remote debugging).

Provides an MCP server wrapper/integration for xlwings, enabling an agent to interact with Excel (via xlwings) through MCP tools.

Agent-guardrails is a shell-based toolkit that installs and wires mechanical enforcement for AI coding agents via git pre-commit hooks and local validation scripts. It helps prevent common bypass patterns and hardcoded secret leaks, and encourages an import-based “registry” pattern (via a project __init__.py template) so new code imports validated functions rather than reimplementing them.

Askimo is a local-first AI desktop app and CLI that connects users to multiple LLM providers (cloud and local), supports persistent chat sessions stored locally (SQLite), performs document/code search with hybrid RAG (BM25 + vector), integrates MCP tools (via stdio or HTTP), and can run scripts (Python/Bash/JavaScript) from the chat. It claims local-only telemetry (token/cost/RAG performance) with data kept on disk.

Cadre-AI is a voice-controlled AI agent that streams audio via WebSockets to a FastAPI server, uses Gemini Live for real-time speech and tool orchestration, and executes actions through MCP toolsets (financial data, web search/weather, and optionally Revit integration via a named-pipe proxy). It supports a local mode (including Revit when enabled) and a Cloud Run mode (Revit disabled).