tech-search

Automate deep technical research with parallel web search and structured report generation.

Updated May 4, 2026
One-click install
npx skills add https://github.com/hitoshiseki/jutsu-simulator --skill tech-search-hitoshiseki
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tech-search
Source: https://github.com/hitoshiseki/jutsu-simulator/tree/main/.claude/skills/tech-search
Command: npx skills add https://github.com/hitoshiseki/jutsu-simulator --skill tech-search-hitoshiseki

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates the hours of manual effort required for deep technical research by automating the full end-to-end pipeline from query parsing to synthesized, documented findings, with no external setup or dependencies needed.

Core Features & Use Cases

  • 6-Phase Automated Workflow: Handles auto-clarification of queries, decomposition into searchable sub-queries, parallel web search via Haiku workers, coverage evaluation, finding synthesis, and structured report generation.
  • Flexible Search Tooling: Works with optional MCP tools like Exa and Context7 for enhanced search and documentation retrieval, with automatic fallbacks to built-in web search and fetch tools if MCPs are unavailable.
  • Strict Scope Enforcement: Built-in guardrails prevent unauthorized file writes, code implementation, or creation of production artifacts, ensuring all output is limited to designated research documentation folders.
  • Use Case: If you need to compare the tradeoffs of React Server Components vs Client Components, this Skill automatically runs parallel searches across credible sources, evaluates coverage, and saves a complete research report with recommendations to your docs folder.

Quick Start

Use the tech-search skill to run a deep research report on any technical topic of your choice, such as "React Server Components vs Client Components".

Frequently Asked Questions about tech-search

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate deep technical research and generate a structured report?

Automating deep technical research is done by decomposing queries into sub-queries, running parallel web searches, scoring source credibility, and synthesizing findings into a structured report, eliminating manual effort.

What is the best way to compare software libraries and evaluate their tradeoffs?

Comparing library tradeoffs is handled by running parallel web searches across credible sources, evaluating coverage of the findings, and saving a complete research report with recommendations to your docs folder.

Do I need external dependencies or MCP tools to run parallel web searches?

No external dependencies are required to run parallel web searches. The workflow uses built-in web search and fetch tools with automatic fallbacks, though optional MCP tools like Exa and Context7 can enhance retrieval.

Can I use this automated research workflow for AI/ML and infrastructure domains?

Yes, automated research workflows apply directly to AI/ML and infrastructure domains, handling state-of-the-art research scenarios, technical comparisons, and library evaluations across software engineering contexts.

What are the limitations of using automated query decomposition for technical research?

Limitations include strict scope enforcement that prevents unauthorized file writes, code implementation, or creation of production artifacts, ensuring all output is limited to designated research documentation folders.