tinker-from-docs
Fine-tune LLMs using the Tinker API. Covers supervised fine-tuning, reinforcement learning, LoRA training, vision-language models, and both high-level Cookbook patterns and low-level API usage.
Найдите подходящую возможность для вашего агента.
Fine-tune LLMs using the Tinker API. Covers supervised fine-tuning, reinforcement learning, LoRA training, vision-language models, and both high-level Cookbook patterns and low-level API usage.
Entry point for Codex-discoverable skills used by the Run-Smart AI coach.
Use this whenever a user wants to add a new feature or explitly states to research a feature/API or building a plan for a new feature. It iterviews the user for feature details (if not provided), research the best API/service for their needs, confirm choice, then gather all implementation notes for their request and save them as a .claude/plans file.
AI thinking partner for strategy and planning. Reads the current work state, analyzes progress, and helps you plan the next phase while the worker instance builds. Triggered by "think ahead".
Local cloud server providing API access to skills and files for Claude.ai
Vercel AI SDK v5: generateText, streamText, useChat, tool calling (use tool() helper + inputSchema), embeddings. v5: sendMessage vs append, message.parts, string models.
Assists in writing ebooks 2-8 of The Quantum Templ™ System, maintaining consistency with Gate I's established voice, structure, and philosophy. Use when creating new Gate ebooks, drafting sections, or ensuring stylistic continuity across the 8-Gate system. Triggers on requests involving Gate II through Gate VIII content creation, ebook drafting, or Quantum Templ writing assistance.
Enhanced project documentation with AI-powered features and Context7 integration
Use when integrating LLMs (OpenAI, Qwen, Claude), extracting structured data from text, building prompts, parsing AI responses, handling JSON output, or implementing multi-step AI workflows
Audit and improve Claude Code configuration. Use when user asks to 'check settings', 'audit config', 'fix settings', 'update claude config', or mentions settings.json issues.
Ensure Claude Code skills meet quality standards through validation operations for structure, content, patterns, and production readiness. Task-based validation with pass/fail criteria, automated checks, and compliance reporting. Use when validating skills before deployment, ensuring standards compliance, certifying production readiness, or quality gating skill releases.
Create event-driven hooks for Claude Code plugins (PreToolUse, PostToolUse, Stop, SessionStart, etc.). WHEN: MUST use when creating Claude Code hooks. Invoke with "/claude-hook-development" or "create a hook", "new PreToolUse hook", "write PostToolUse handler". WHEN NOT: Creating agents (use claude-agent-creator), writing hookify rules (use claude-hookify-rules).
Generate automation scenarios from natural-language requests by drafting `scenarios/<name>/plan.json` plus supporting files (behaviors, curl, SQL mocks). Use when a user asks to create or update a scenario or wants a plan built from requirements in plain Portuguese.
Native multi-agent orchestration using Claude Code's Agent Tool. Use when multiple independent tasks can run with different domain expertise or when comprehensive analysis requires multiple perspectives.
Automated coordination, formatting, and learning from Claude Code operations using intelligent hooks with MCP integration. Includes pre/post task hooks, session management, Git integration, memory coordination, and neural pattern training for enhanced development workflows.
Oracle Cloud Infrastructure expertise for enterprise AI and cloud architecture
Transformer architecture fundamentals. Covers self-attention mechanism, multi-head attention, feed-forward networks, layer normalization, and residual connections. Essential concepts for understanding LLMs.
Expert in load balancing and dynamic task allocation for multi-agent systems. Specializes in optimal routing based on agent capability, availability, and cost (Token Economics).