আপনার এজেন্টের জন্য উপযুক্ত সক্ষমতা খুঁজে নিন।
Propose multi-agent and workflow architectures for CustomGPT.ai Labs projects, showing how Claude Code, CustomGPT.ai, and supporting tools interact to deliver the desired business outcome.
Guide for building AI-powered applications using the Vercel AI SDK v6. Use when developing with generateText, streamText, useChat, tool calling, agents, structured output generation, MCP integration, or any LLM-powered features in TypeScript/JavaScript applications. Covers React, Next.js, Vue, Svelte, and Node.js implementations.
Ralph Loop integration patterns for iterative AI development. Use when starting Ralph loops, managing iterations, or recovering from context compaction. Covers TDD, debugging, context management, and memory persistence.
AI 모델 API 호출명 및 가격 참조 가이드. API 키로 AI 모델을 호출할 때 정확한 모델명(model string)과 최신 가격 정보를 제공합니다. 사용 시점: (1) OpenAI, Anthropic, Google, DeepSeek 등의 API 호출 시 모델명이 필요할 때, (2) 토큰 비용/가격 비교가 필요할 때, (3) 최신 추론 모델/FAST 모델/가성비 모델 선택이 필요할 때, (4) 프롬프트 캐싱/배치 처리 비용 최적화가 필요할 때
Automatically applies when creating AI tool functions. Ensures proper schema design, input validation, error handling, context access, and comprehensive testing.
Configure ADK bidi-streaming for real-time multimodal interactions. Use when building live voice/video agents, implementing real-time streaming, configuring LiveRequestQueue, setting up audio/video processing, or when user mentions bidi-streaming, real-time agents, streaming tools, multimodal streaming, or Gemini Live API.
Persistence and human-in-the-loop patterns for LangGraph. Use when implementing checkpointing, saving/restoring graph state, adding human approval workflows, managing conversation memory across sessions, enabling time-travel debugging, or building fault-tolerant agents that resume from failures.
Comprehensive research and synthesis agent specializing in multi-source information gathering, critical analysis, and integrated knowledge synthesis. Excels at complex research projects requiring systematic investigation across domains, evidence evaluation, and coherent narrative construction.
This skill should be used when the user asks about "Hugging Face", "HF Hub", "transformers", "model hub", or needs guidance on which Hugging Face capability to use. Acts as an entry-point that routes to specialized HF skills (cli, jobs, datasets, evaluation, model-trainer, paper-publisher, trackio, tool-builder) based on the task. Use for authentication setup, quick operations, and choosing the right specialized skill.
Generate optimized indirect prompt injection, H-CoT, and multi-layer attack payloads for AI security testing and CTF competitions with automated family selection and success rate optimization
A two-phase repair skill that analyzes errors and suggests fixes before executing repairs. Phase one: user describes error, agent analyzes and proposes solution. Phase two: upon approval, executes the repair action.
Use when building AI agents with Microsoft Agent Framework (Semantic Kernel + AutoGen unified); when implementing memory or context providers; when threads won't deserialize; when workflow checkpointing fails; when migrating from Semantic Kernel or AutoGen; when seeing ChatAgent or AgentThread errors
Documentation for vercel-ai-sdk. Use for questions about vercel-ai-sdk APIs, patterns, and usage.
Expert guidelines for Python ML and LLM workflows. Covers code quality, experiment tracking, and data handling. Use when working on AI/ML components or data pipelines.
Build Retrieval Augmented Generation (RAG) systems for AI applications. Use when creating document Q&A systems, knowledge bases, semantic search, or any application combining retrieval with LLM generation. Triggers include "RAG", "vector database", "embeddings", "document chunking", "semantic search", or "knowledge base".
Use for any TiDB-related projects, tasks, or code. Context engineering and context management for AI agents: keep prompt prefixes stable for KV-cache, use append-only context, prefer tool masking over tool removal, offload large observations into filesystem memory, recite goals/todos to control attention, preserve errors for recovery, and avoid few-shot pattern lock-in. Use when building or debugging agent loops, prompt/context schemas, memory strategies, or tool-availability policies.
