ai-toolkit

Plugin ai-toolkit from marketplace zAI-Skills in Aznatkoiny/zAI-Skills. Commands read from the repo's manifest — 5 days ago.

Skills (6)

• cpp-reinforcement-learning

  Reinforcement learning in C++ with LibTorch (the PyTorch C++ frontend) and C++20. Use when the user is implementing DQN, PPO or SAC in C++, building a replay buffer or rollout buffer in C++, fixing libtorch memory growth or GPU issues in a C++ agent, running environments in parallel threads, or load

• deep-learning

  Build and train neural networks with Keras 3 on any backend (JAX, TensorFlow or PyTorch). Use when the user is writing Keras code or asks for a Keras model, wants to migrate tf.keras / Keras 2 code to Keras 3, or is working on image classification, segmentation, text classification, sequence-to-sequ

• openclaw-setup

  Install, configure, secure and troubleshoot OpenClaw (formerly Clawdbot / Moltbot), the self-hosted personal AI assistant that answers through messaging apps. Covers installing on a Mac or Linux machine or a VPS with Docker (including Hostinger), Anthropic API setup and cost control, Telegram and iM

• prompt-engineer

  Write, review, debug, and migrate prompts for Claude models (Fable 5.1, Opus 5.5, Sonnet 5.5, Haiku 5.5) using Anthropic's official prompt engineering guidance. Covers system prompts, agent and harness prompts, tool descriptions, few-shot examples, and effort/thinking settings. Use this whenever the

• reinforcement-learning

  Reinforcement learning in Python with Gymnasium and Stable-Baselines3 (PPO, SAC, DQN, TD3, A2C). Use when the user is building a custom Gymnasium environment, choosing an algorithm, training or tuning an agent, debugging an agent that won't learn (flat reward, collapse, NaNs), evaluating a policy, o

• x402-payments

  Build apps that charge for, or pay for, HTTP requests with the x402 protocol (HTTP 402 "Payment Required" plus stablecoin payments, usually USDC on Base or Solana). Covers paid API endpoints and middleware (Express, Hono, Next.js), clients and AI agents that pay for x402 resources with spending limi

Slash commands (4)

• /ai-toolkit:openclaw-audit

  Read-only security check of an existing OpenClaw install (native or Docker) — version, exposure, auth, channel policies, tool limits — with a pass/fail table and fixes

• /ai-toolkit:optimize-prompt

  Review and rewrite a prompt (system prompt, agent prompt, skill, CLAUDE.md, tool description) for current Claude models — returns the rewritten prompt, what changed and why, settings when the API is involved, and how to test it

• /ai-toolkit:rl-new-project

  Scaffold a Python RL project — custom Gymnasium environment, Stable-Baselines3 training with evaluation and VecNormalize, an Optuna tuning stub, and Makefile targets that validate the environment before training

• /ai-toolkit:x402-scaffold

  Generate a working x402 starter on testnet — a seller (an endpoint that charges per request) or a buyer (a client that pays) — on EVM (Base) or Solana

Subagents (2)

• prompt-rewriter

  Rewrites one long or high-stakes Claude prompt in a clean context using the bundled prompt-engineer skill: long system prompts, agent or harness prompts, prompts being migrated to a newer Claude model. Returns the rewritten prompt, what changed and why, settings when the API is involved, and how to

• rl-debugger

  Diagnoses why a Python reinforcement-learning run (Gymnasium + Stable-Baselines3) isn't working: flat or collapsing reward, entropy collapse, value-function trouble, NaN losses, runs that vary wildly across seeds, or an agent that trains well but fails in evaluation. Reads the code and logs, runs th

Similar plugins