How do I trigger the research skill?

The comprehensive research workflow is specifically designed to be invoked via the /ai-eng/research command within your AI-augmented engineering environment.

Does it provide specific code references?

Yes, every finding includes an evidence chain with specific file paths and line-number ranges to ensure all insights are verifiable against the source code.

How does the research confidence score work?

After completing an investigation, the skill rates findings from 0.0 to 1.0 based on evidence completeness, historical context, and the clarity of the conclusions.

Can this skill search the web?

Yes, for complex research tasks, the skill can conditionally spawn a web-search-researcher agent when external context or third-party documentation is needed.

What is the benefit of parallel discovery?

Parallel discovery allows multiple specialized agents (codebase, documentation, and pattern finders) to scan your project simultaneously, maximizing coverage while minimizing wait time.

Comprehensive Research

Name: Comprehensive Research
Author: v1truv1us

byv1truv1us

•

Productivity & Workflow

Orchestrates multi-phase investigations across codebases, documentation, and external sources to generate evidence-based insights.

About

Comprehensive Research is a sophisticated orchestration skill that automates the investigation of complex software systems through a structured four-phase methodology: scope definition, parallel discovery, sequential analysis, and synthesis. By coordinating specialized agents, it provides deep codebase insights, identifies architectural patterns, and extracts historical decision-making context. This skill is essential for developers who need to ground their implementation decisions in thorough evidence, avoid repeating historical mistakes, and produce authoritative documentation without getting lost in the noise of a large project.

Key Features

Conditional integration with external web search and domain specialists
Multi-phase orchestration (Discovery, Analysis, and Synthesis)
2 GitHub stars
Automated research confidence scoring and uncertainty identification
Evidence-based reporting with specific file and line-number references
Parallel agent execution for rapid codebase and documentation scanning

Use Cases

Synthesizing disparate information from source code, local docs, and external knowledge
Investigating legacy codebases to understand historical design decisions and constraints
Conducting deep impact analysis before implementing major architectural changes

About

Key Features

Conditional integration with external web search and domain specialists
Multi-phase orchestration (Discovery, Analysis, and Synthesis)
2 GitHub stars
Automated research confidence scoring and uncertainty identification
Evidence-based reporting with specific file and line-number references
Parallel agent execution for rapid codebase and documentation scanning

Use Cases

Synthesizing disparate information from source code, local docs, and external knowledge
Investigating legacy codebases to understand historical design decisions and constraints
Conducting deep impact analysis before implementing major architectural changes