Morningstar202604

Claude Tool Agent — AI skill for Claude Code

AI community

An interactive LLM tool-calling agent built on Claude's native tool-use API and FastAPI.

How to install Claude Tool Agent

This entry records only its repository, not the path inside it, so there is no exact command to give. Open Morningstar202604/claude-tool-agent and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Claude Tool Agent does

An interactive LLM tool-calling agent built on Claude's native tool-use API and FastAPI. Streams text, tool calls, and tool results to the browser over SSE while Claude orchestrates multiple tools across turns. Ships with web search, Python code execution, Wikipedia lookup, and sandboxed file read tools out of the box, plus a side-by-side Lang

Alternatives in AI

Cursor Tools — by eastlondoner - Creates a versatile AI command interface supporting multiple providers and models with flexi 4.7k ★
Vllm Mlx — High-performance OpenAI and Anthropic compatible LLM inference server for Apple Silicon 1.5k ★
Claw Empire — Command Your AI Agent Empire from the CEO Desk — A local-first AI agent office simulator that orchestrates CLI 1.1k ★

README

AI 群聊 — WeChat-style Multi-Room Chatroom

像微信一样建群、拉 AI 角色进群、写群公告，然后一起聊天。
Quick Start · Features · Architecture · Usage
Languages: English (current) · 中文

Overview

**AI 群聊** turns a single AI chat into a **WeChat-style multi-room group chat platform**. Create multiple chat groups, each with its own name, roster of AI roles (pulled from an address book or created on the fly), group announcement, and conversation history — all persisted in your browser.

Left sidebar: group list with search — switch between rooms instantly
Room management: create, edit, and delete groups; each group has its own roles, announcement, rules, and chat history
Contact library: preset AI role definitions (研究员 / 代码手 / 创意家 / 评论家 / 编剧 / 历史学家) that you can pull into any group
Custom roles per room: one-off roles that belong only to a specific group
Group announcement + rules: the announcement is shown as a banner at the top of the chat and prepended to every AI role's system prompt; supplementary rules go after it
@Mention chaining: type @角色名 to direct specific roles to speak next
SSE token streaming: every AI reply streams token-by-token with a blinking cursor
Stateless backend: all room config lives in `loc