search▌

Similarity Search Patterns \n Patterns for implementing efficient similarity search in production systems. \n Use this skill when \n \n Building semantic search systems \n Implementing RAG retrieval \n Creating recommendation engines \n Optimizing search latency \n Scaling to millions of vectors \n Combining semantic and keyword search \n \n Do not use this skill when \n \n The task is unrelated to similarity search patterns \n You need a different domain or tool outside this scope \n \n Instruc

Parallel multi-topic research with independent agents generating structured knowledge bases from any input material. \n \n Launches dedicated research agents for each topic in parallel, each conducting 4–6 rounds of deep retrieval with deduplication checks to avoid keyword overlap \n Accepts three input modes: file-based ( /multi-search @path.md ), direct paste, or explicit topic specification with custom research directions \n Generates structured output with a research overview document, indiv

If you see unfamiliar placeholders or need to check which tools are connected, see CONNECTORS.md.

Semantic search across web, research papers, and similar content discovery via Exa API. \n \n Five endpoint modes: semantic/keyword search, content extraction, similar page discovery, direct answers, and structured research with custom output schemas \n Supports search type selection (neural embeddings, fast keyword, deep comprehensive, or auto), category filtering (research papers, GitHub, news, PDFs, tweets, etc.), and date range constraints \n Two-phase architecture with main skill for intent

Search persistent cross-session memory to answer questions about previous work and past solutions. \n \n Three-layer workflow: search for IDs and metadata (50–100 tokens per result), optionally view timeline context around interesting results, then batch-fetch full details only for filtered IDs \n Search parameters include query text, project name, observation type (bugfix, feature, decision, discovery, change), and date range filtering \n Timeline tool shows observations, sessions, and prompts

Five AI-optimized web tools for search, content extraction, site crawling, mapping, and research reports. \n \n Five distinct tools: tavily_search for web queries with AI answers, tavily_extract for clean content from URLs, tavily_crawl for multi-page site traversal, tavily_map for URL discovery, and tavily_research for comprehensive multi-step research \n Supports advanced search modes including deep research, topic filtering (general/news/finance), and time-range constraints \n Crawl and map o

Neural search for web content, code, companies, and people via the Exa MCP server.

Fast web search for current information, research, and fact-finding across the internet. \n \n Executes single objective-based queries or multiple keyword searches in parallel, returning up to 10 results with excerpts and metadata \n Supports time-sensitive filtering via --after-date and domain-specific searches with --include-domains \n Outputs structured JSON with titles, URLs, publish dates, and excerpts for easy parsing and follow-up queries \n Requires inline citations for every claim using

Modern React InstantSearch setup using hooks for type-ahead search.

Lightweight web search and content extraction via Brave Search API, no browser required. \n \n Two core commands: search.js for querying the web with configurable result counts, and content.js for extracting readable markdown from specific URLs \n Search results include title, link, and snippet by default; add --content flag to include full page content as markdown \n Ideal for documentation lookup, fact-checking, and fetching web content without the overhead of browser automation \n Requires BR