Найдите подходящую возможность для вашего агента.
Meta-orchestration guide for choosing optimal reasoning patterns. Analyzes problem characteristics and recommends which cognitive methodology to use - tree-of-thoughts (find best), breadth-of-thought (explore all), self-reflecting-chain (sequential logic), or direct analysis. Use when facing complex problems and unsure which reasoning approach fits best.
Fine-tune LLMs with Unsloth using GRPO or SFT. Supports FP8, vision models, mobile deployment, Docker, packing, GGUF export. Use when: train with GRPO, fine-tune, reward functions, SFT training, FP8 training, vision fine-tuning, phone deployment, docker training, packing, export to GGUF.
How to track our work through commits, pull-requests, and pushes. Maintain proper git hygiene in our chaotic, multi-agent environment through atomic commits, with the option to have code reviewed by AI agents with PRs.
Test and verify AI orchestrator features using CLI runner, unit tests, and E2E tests. ALWAYS use this skill when: implementing new features, fixing bugs, making changes to agents/orchestrator/TUI, or when asked to verify something works. Claude MUST test changes before marking tasks complete.
Parallel execution patterns for cognitive reasoning tasks. Use when independent sub-problems can be solved simultaneously, multiple solution approaches need exploration, ensemble confidence is required, or time permits depth without sequential constraints. Integrates with ToT, BoT, HE, and AT for accelerated reasoning with fan-out/fan-in, MCTS-style search, and MoA aggregation patterns.
Debug detection and routing misclassifications in OpenEvent-AI (LLM vs regex, qna_types, confirmations, OOC drops, site-visit triggers). Use when intent seems wrong, Q&A is ignored, confirmations route to the wrong step, or site visit/date change triggers unexpectedly.
Export Claude skills. Use to generate documentation of all available Claude Skills.
Create new Claude Code agent skills. Automatically analyzes requirements and generates complete skill with proper structure, triggers, and tools. Use when you want to "create a skill", "build a new skill", "make a skill for", or automate repetitive tasks.
Generate complete, production-ready Claude Code Skills based on business domain and use cases. Use this skill when you want to CREATE NEW SKILLS, BUILD SKILL PACKAGES, or GENERATE AUTOMATION TOOLS for any business domain. Creates self-contained skill packages with SKILL.md, Python scripts, test data, and sample prompts following best practices.
Transform raw, unorganized information into actionable knowledge through systematic validation. Use when users want to synthesize information from multiple sources (documents, URLs, transcripts, notes) into structured knowledge documents. Supports three document types - curated context (default, optimized for recall), guidance (implementation-focused narrative), and reference (quick lookup). Combines convergent synthesis with tension preservation to maintain productive contradictions. Triggers on requests like "synthesize this information", "create knowledge document from these sources", "transform these notes into actionable guidance", or "help me organize this research".
Run OpenAI Codex CLI reviews from Claude Code. Use when the user asks for a Codex review or mentions triggers like "codex-review", "/codex-review", or "codex review".
Creating and Optimizing Claude Code Skills. Design reusable knowledge capsules with progressive disclosure (metadata → content → resources). Apply freedom levels (high/medium/low), create examples, validate YAML. Use when building domain-specific guidance or automating recurring patterns.
Deploy and manage Lark (Feishu) bot for Milady meme generation, command handling, and team collaboration. Use when setting up Lark webhooks, handling bot commands, deploying to production, or integrating with Lark messaging.
Reconnaissance techniques for exploring large contexts before reading them. Teaches agents to probe, filter, and map data structures without consuming context window.
Recognize, diagnose, and mitigate patterns of context degradation in agent systems. Use when context grows large, agent performance degrades unexpectedly, or debugging agent failures.
SmartACE (Agentic Context Engineering) workflow engine with MCP-B (Master Client Bridge) and AMUM-QCI-ETHIC module. Dual database architecture using DuckDB (analytics) + SurrealDB (graph). Uses Blender 5.0 (bpy) and UE5 Remote Control. Use when (1) MCP-B agent-to-agent communication (INQC protocol), (2) AMUM 3→6→9 progressive alignment, (3) QCI quantum coherence states, (4) ETHIC principles enforcement (Marcel/Anthropic/EU AI Act), (5) SurrealDB graph relationships, (6) DuckDB SQL workflows, (7) ML inference with infera/vss, (8) Blender 5.0 headless processing, (9) UE5 scene control, (10) DuckLake time travel.
Create Claude Code subagents. Use when user wants to create a subagent, specialized agent, or custom AI assistant for Claude Code.
Enterprise Mermaid diagramming skill for Claude Code using MCP Playwright
