chrome-devtools
Chrome DevTools for AI agents. Control and inspect live Chrome browsers with performance tracing, network inspection, DOM snapshots, and automated interactions. Use for web debugging, performance analysis, and...
发现 tools llm ai 相关 Skills,比较来源仓库和用途,查看适合你工作流的技能说明。
Chrome DevTools for AI agents. Control and inspect live Chrome browsers with performance tracing, network inspection, DOM snapshots, and automated interactions. Use for web debugging, performance analysis, and...
OpenAI API for chat completions, embeddings, and assistants
Generates architecture, database, and system diagrams using Mermaid syntax. Creates visual representations of system architecture, database schemas, component relationships, and data flows.
Pinecone vector database for similarity search and RAG
Reddit API for market research and sentiment analysis
AI-powered code review using Gemini models
Vercel AI SDK agent skill management
Hugging Face Hub integration for model and dataset operations
Multi-agent conflict resolution protocol for detecting and resolving conflicts when multiple agents produce conflicting outputs or requirements. Handles technical, requirements, design, and data conflicts.
LangChain framework utilities for chains, agents, and RAG
Classify code, documents, and data into categories. Use for code categorization, content classification, and data organization.
Convert natural language queries to SQL. Use for database queries, data analysis, and reporting.
Manages memory tool integration with CLAUDE.md files for dual persistence and redundancy. Handles cross-conversation learning, memory sync, and context persistence.
User persona management and profile generation
Agent scaffolding and template generation for Vercel deployments
Automatically generates claude.md files for new folders/modules following hierarchical structure. Extracts context from existing code, follows project conventions, and creates documentation that enables Claude...
Persistent knowledge graph memory system. Store and retrieve entities, relations, and observations across conversations. Use for maintaining user context, preferences, and learned patterns.
Synchronize task state and metadata across Claude and Cursor sessions. Use when handing off tasks between platforms, sharing plans, or updating external trackers like Linear or Jira.
基于 LLM 的深度迭代搜索与推理服务。擅长处理复杂问题,通过自动分解查询、多轮迭代检索、信息评估与验证,最终生成全面且结构化的深度分析报告。
文本向量化(Embedding)基础服务。将自然语言转换为高维稠密向量,为语义搜索、聚类分析、推荐系统等下游任务提供核心数据支持。
高性能 RAG 多路检索服务。集成 Milvus 向量数据库进行语义检索,并结合 Rerank 模型进行精准重排序,支持海量文档的高效存储与历史内容召回。
Use when getting started with llmemory document storage and search - covers installation, initialization, adding documents, vector search, hybrid search, semantic search, BM25 full-text search, document managem...
Use when building search systems that need both semantic similarity and keyword matching - covers combining vector and BM25 search with Reciprocal Rank Fusion, alpha tuning for search weight control, and optimi...
Use when search queries need better recall through query expansion - generates multiple query variants, retrieves with each, and fuses results using RRF for improved retrieval quality especially with ambiguous...
默认按已收录数据中的 GitHub 仓库 Stars 排序。Stars 属于来源仓库,同一仓库的多个技能可能共享相同数值;它反映仓库关注度,不代表单个技能的安装量或质量评分。