skills.rest
Sanery1
Sanery1COMMUNITY
Explore all skills by this author →
WebsiteGitHub

FitClaw-old

Integrate multi-vendor LLMs with ReAct tool cycles and event streams.

Updated Apr 29, 2026
Search Tags:#LLM#AI Agent#sports coaching#smart programming#multi-vendor integration
One-click install
npx skills add https://github.com/Sanery1/FitClaw-old --skill fitclaw-old
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: FitClaw-old
Source: https://github.com/Sanery1/FitClaw-old/tree/main/packages/coding-agent/test/fixtures/skills/root-skill-preferred
Command: npx skills add https://github.com/Sanery1/FitClaw-old --skill fitclaw-old

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires typescript, npm, node, python, docker, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

FitClaw-old solves the problem of manual exercise planning and coding, providing a smart programming assistant to assist with tasks ranging from model integration to user interaction.

Core Features & Use Cases

  • Agent Runtime: Unifies model calls, tool execution, state management, and event systems.
  • CLI Programming Assistant: Provides an interactive development experience similar to Claude Code.
  • Feishu Bot: Integrates Agent capabilities into instant messaging and team workflows.
  • Sports Coach Skill: Offers training plans, motion query, training record, and physical examination management.
  • Web/UI Components: Accumulates reusable chat interface and terminal UI capabilities.
  • GPU Pod Management: Assists in the deployment and management of local/remote inference resources such as vLLM.

Quick Start

Start by installing dependencies and then building all packages using npm.

  • Install dependencies: npm install
  • Build all packages: npm run build

To start the CLI, run the command: node packages/coding-agent/dist/cli.js

Frequently Asked Questions about FitClaw-old

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a full-stack AI Agent with multi-vendor LLM integration and a ReAct tool cycle?▼

To build a full-stack AI Agent with multi-vendor LLM integration, you can use this platform to unify model calls, tool execution, and state management through its built-in ReAct tool cycle and event stream.

Can I integrate an AI Agent into team workflows using a Feishu Bot?▼

Yes, you can integrate AI Agent capabilities into team workflows using a Feishu Bot, which embeds the agent's interactive features directly into instant messaging for collaborative development.

How do I set up the CLI programming assistant for an interactive development experience?▼

To set up the CLI programming assistant, install dependencies with npm install, run npm run build, and execute node packages/coding-agent/dist/cli.js to start the interactive terminal interface.

Does this AI Agent platform require TypeScript and Docker to manage GPU Pods?▼

Yes, managing GPU Pods for local or remote inference resources like vLLM requires Docker, while the agent runtime and CLI programming assistant rely on TypeScript and npm for package building.

What is the best way to manage training plans and physical examinations using an AI Agent?▼

The best way to manage training plans and physical examinations is using the integrated sports coach skill, which offers motion queries and training record management within the AI agent platform.

How does the Agent Runtime handle state management and event systems?▼

The Agent Runtime handles state management and event systems by unifying model calls and tool execution into a single stream, ensuring consistent event flow across CLI, Web UI, and Feishu Bot interfaces.

Related Skills

Explore More Software Engineering Skills→
CommonCapitalCommonCapital
2

nodejs-sandbox

Execute Node.js and TypeScript code in a Docker container.

Community
Intermediate
liutao96liutao96

lark-event

Subscribe to Feishu events via WebSocket and emit NDJSON to stdout.

Community
Intermediate
Zangetsu101Zangetsu101
1

prototype

Create interactive terminal and UI prototypes for validating design and logic.

Community
Advanced
SignmanalSignmanal
1

openhands

Delegate coding tasks with file edits and shell commands to a LiteLLM-compatible agent via CLI.

Community
Intermediate