research

Research technical topics and generate markdown reports with actionable recommendations.

1|Updated Dec 10, 2025
One-click install
npx skills add https://github.com/hungtrandigital/splt --skill research-hungtrandigital
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research
Source: https://github.com/hungtrandigital/splt/tree/main/0-agents/agents/skills/research
Command: npx skills add https://github.com/hungtrandigital/splt --skill research-hungtrandigital

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill speeds up technical research and planning by providing a rigorous, phased approach to scope, gather information, evaluate sources, and synthesize action-oriented recommendations.

Core Features & Use Cases

  • Structured research phases: Scope definition, systematic information gathering, analysis, and synthesis aligned with YAGNI, KISS, and DRY principles.
  • Multi-source strategy: When possible, use Gemini for searches (gemini -m gemini-2.5-flash -p "<topic>") and save reports under ./plans/<plan-name>/reports/YYMMDD-<topic>.md; fallback to WebSearch when Gemini isn't available.
  • Actionable outputs: Executive summaries, implementation plans, and risk/mitigation insights suitable for planning scalable solutions.

Quick Start

Start a research plan for "<topic>" and, if Gemini is available, run a search with gemini -m gemini-2.5-flash -p "<topic>" (timeout: 10 minutes) and save to ./plans/<plan-name>/reports/YYMMDD-<topic>.md. If Gemini isn't available, use the WebSearch tool to gather sources, then proceed to Phase 3 and 4.

Frequently Asked Questions about research

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

FAQPage Schema
How do I plan scalable and secure technical solutions systematically?

Scalable, secure solutions require phased planning: scope definition, multi-source information gathering, source evaluation, and synthesis into actionable recommendations. This Skill provides structured research phases aligned with YAGNI, KISS, and DRY principles to ensure verifiable, current sources and reproducible outputs.

What's the best way to gather and synthesize technical information from multiple sources?

Multi-source technical research uses systematic information gathering, leveraging Gemini for searches when available or WebSearch as fallback, then evaluates and synthesizes findings into markdown reports. This approach ensures currency within 12 months and bounded tool usage for actionable insights.

How do I create structured technical research reports for architecture and security decisions?

Generate structured markdown reports by following a phased workflow: define scope, gather information across multiple sources, analyze findings against technical requirements, and synthesize executive summaries with implementation plans and risk mitigation strategies suitable for architecture and operations contexts.

Can I use this for research on software architecture, performance, and operations together?

Yes. This Skill applies across software architecture, security, performance, and operations contexts. It satisfies functional and technical requirements for all these domains, delivering verifiable recommendations and reproducible outputs from bounded research phases.

What sources and timeframes does technical research require to be actionable?

Actionable technical research requires verifiable sources with currency within the last 12 months, structured markdown outputs, and reproducibility. The Skill enforces these constraints through phased information gathering and synthesis to support planning and decision-making.

When should I use Gemini versus WebSearch for technical research?

Use Gemini with `gemini -m gemini-2.5-flash` for primary searches (10-minute timeout), saving reports to `./plans/<plan-name>/reports/`. Fall back to WebSearch only when Gemini is unavailable, then proceed to analysis and synthesis phases with the gathered sources.