slurpyb/searching-web-and-files

Apply precise writing style that expresses ideas clearly with personal voice and logical flow.

Prioritize brevity, directness, and clarity in all responses with minimal token usage.

Subject-neutral deep research workflow. Decomposes any topic into facet axes, runs 6+ parallel fan-out rounds across mixed sources (web/registry/repo/docs), produces datestamped append-only artifacts, and synthesizes a schema-validated report with a full citation index. Use when the user wants to "research deeply", "become an expert on", "scrape and synthesize", "deep dive", "investigate", or "build a knowledge base on" a topic.

Build hands-on, code-first tutorials from any primary source - technical documentation, codebases, APIs, tools, or other complex material. Extract real examples, working code, and concrete scenarios. Create tutorials using markdown (text-heavy summaries) or React artifacts (complex interactive workflows). Keywords - tutorial, codebase, API, hands-on, code-first, copy-paste, interactive, real examples, primary source

Neural search via Exa MCP for web, code, and company research. Use when the user needs web search, code examples, company intel, people lookup, or AI-powered deep research with Exa's neural search engine.

Explore complex, multi-faceted topics requiring deep understanding through a three-round iterative strategy with parallel subagents.

Write content to structured .txt files using markdown and XML formatting.

Explain solutions through the lens of discovery and conversation journey, emphasizing how insights emerged rather than exhaustive technical details. Use when explaining code changes, decision-making processes, problem-solving approaches, or any narrative that benefits from showing the "why" before the "what". Trigger with JOURNEY keyword.

Automatically check and update folder-specific AGENTS.md during research. Before investigating a domain, read nearest AGENTS.md for existing context. After discovering valuable patterns, append learnings to that file.

Investigation flow (Glob -> Grep -> Read), evidence-based research with file:line references, structured output format for AI consumption. Use for pattern discovery, implementation research, and codebase investigation.

Structured research summarization agent skill for non-dev users. Handles academic papers, web articles, reports, and documentation. Extracts key findings, generates comparative analyses, and produces properly formatted citations. Use when: user wants to summarize a research paper, compare multiple sources, extract citations from documents, or create structured research briefs. Plugin for Claude Code, Codex, Gemini CLI, and OpenClaw.

Synthesize YouTube and competitor research into content gaps, opportunity scores, and actionable idea briefs. Use after gathering video metadata, transcripts, comments, or competitor evidence.

Research-before-coding workflow. Search for existing tools, libraries, and patterns before writing custom code. Invokes the researcher agent.

Use BEFORE any search — locating code, files, symbols, docs, prior art, or researching a topic — whenever you need to find something and don't already know its exact location. Casts wide in parallel, learns from each round, and keeps iterating instead of grepping one guessed keyword and quitting. Triggers - "find X", "where is Y", "is there a Z", "search for", "look for", "research", "does this exist", or any task that starts by locating things.

Iterative parallel search strategy using WebSearch and Exa. Launch multiple searches simultaneously, learn from failures, and keep iterating until finding answers. Use for research, code examples, documentation, and technical information gathering.

Use for any search that must not miss things — finding code/files/symbols, prior art, docs, or researching a topic where one keyword and a quick read is not good enough. Runs an explicit three-round protocol (cast wide → close gaps → verify & saturate) with parallel searches inside each round and a mandatory synthesis gate between them. Triggers - "find everything", "is there anything that", "thoroughly search", "make sure nothing's missed", "research deeply", "where could X be", or any locate-task where a false "not found" is costly.

Runs wide parallel discovery, written synthesis gates, gap-driven deeper rounds, verification, and saturation across web and local sources. Use when a task requires finding information across the web, documentation, repositories, code, or local files—especially when vocabulary is uncertain, one search may miss results, multiple sources must be synthesized, or a defensible “not found” conclusion is required.

Use this skill for requests related to web research; it provides a structured approach to conducting comprehensive web research

Summarize YouTube videos by extracting and analyzing auto-generated subtitles. Use when the user provides a YouTube URL and asks to summarize, explain, or analyze a video's content. Triggers on patterns like "this video summarize", "summarize this YouTube", "what does this video say", or any request involving a youtube.com or youtu.be URL combined with summarization/analysis intent. Supports both Japanese and English videos.