How does the research confidence assessment work?

It rates the research depth from 0.0 to 1.0 based on evidence coverage and historical context, allowing you to see exactly where evidence is weak or where further investigation is required.

Can this skill search external documentation?

Yes, the orchestrator can trigger a web-search-researcher agent when external context is needed to supplement local codebase information.

Does it support parallel processing?

Yes, the Discovery phase spawns multiple agents simultaneously to scan for files, documentation, and patterns concurrently to maximize efficiency.

What format is the research output delivered in?

It produces a comprehensive, structured Markdown document including a synopsis, detailed findings with code references, historical context, and actionable recommendations.

What makes this different from a standard codebase search?

Unlike basic searching, this skill uses a structured four-phase workflow to cross-reference findings, identify patterns, and synthesize documentation with evidence-based reasoning rather than just returning keyword matches.

Comprehensive Research Orchestrator

Name: Comprehensive Research Orchestrator
Author: v1truv1us

byv1truv1us

•

개발자 도구

Orchestrates multi-phase research agents to conduct deep codebase analysis, documentation discovery, and knowledge synthesis.

The Comprehensive Research skill provides a structured, multi-agent approach to technical investigation within Claude Code. It moves beyond simple file searches by orchestrating specialized agents through parallel discovery and sequential analysis phases, ensuring exhaustive coverage of codebases and documentation. By utilizing evidence-based analysis and a built-in confidence assessment, it helps developers understand complex architectural patterns, historical design decisions, and external technical context before making critical implementation choices.

주요 기능

01Conditional domain specialists for web search, security, and architecture

022 GitHub stars

03Integrated Research Confidence Assessment (0.0 to 1.0) to identify gaps

04Multi-phase orchestration (Discovery, Analysis, Synthesis, and Validation)

05Evidence-based reporting with specific file and line-number references

06Parallel agent execution for rapid codebase and documentation scanning

사용 사례

01Synthesizing scattered documentation and historical context for authoritative reports

02Performing deep analysis on unfamiliar or legacy codebases to identify patterns

03Conducting impact analysis and requirement gathering before major refactors

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add v1truv1us/ai-eng-system comprehensive-research

For use in Claude.ai and ChatGPT

Download Skill