What are Parallel Deep Dives?

This phase deploys multiple specialized agents simultaneously to analyze different parts of a system (e.g., database, API, and frontend) at once, significantly speeding up the research phase.

How does the Investigation Workflow differ from standard coding?

While standard workflows focus on building features, this skill is optimized for deep exploration, using parallel agents to gather information and verify understanding before any code changes are proposed.

How does it help with long-term project knowledge?

It includes a dedicated Knowledge Capture phase that stores discoveries in the project's memory and updates documentation, ensuring that future team members don't have to repeat the same investigation.

When does this skill auto-activate?

It activates on research-oriented prompts such as 'explain the architecture,' 'analyze how this works,' 'understand the system,' or 'investigate the auth flow.'

Can I use this for bug fixing?

It is ideal for the analysis stage of a complex or 'heisenbug' to understand the root cause. Once the cause is identified, the workflow provides a smooth transition to standard development for implementation.

Investigation Workflow Claude Code Skill

The Investigation Workflow skill transforms Claude into a deep-research agent specialized in codebase exploration rather than just implementation. By utilizing a structured 6-phase process—ranging from initial scope definition to parallel agent deployment and final knowledge capture—it helps developers quickly demystify legacy systems, debug architectural bottlenecks, and document system flows. It is optimized for high-efficiency information gathering, targeting a 30-40% reduction in message overhead while ensuring findings are verified through practical testing and stored in project memory for future reference.

Características Principales

01Hypothesis-driven verification phase to validate architectural assumptions

02Parallel agent deployment for simultaneous deep dives into multiple code modules

03Automatic discovery storage in project memory to prevent repeat investigations

0430 GitHub stars

05Systematic 6-phase investigation methodology for structured system analysis

06Synthesis engine that generates executive summaries and visual flow diagrams

Casos de Uso

01Mapping integration flows and dependencies between multiple services

02Deep-dive analysis of persistent system failures or performance bottlenecks

03Onboarding to a large, unfamiliar legacy codebase or complex architecture

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add rysweet/amplihack investigation-workflow

For use in Claude.ai and ChatGPT

Download Skill

Características Principales

01Hypothesis-driven verification phase to validate architectural assumptions

02Parallel agent deployment for simultaneous deep dives into multiple code modules

03Automatic discovery storage in project memory to prevent repeat investigations

0430 GitHub stars

05Systematic 6-phase investigation methodology for structured system analysis

06Synthesis engine that generates executive summaries and visual flow diagrams

Casos de Uso

01Mapping integration flows and dependencies between multiple services

02Deep-dive analysis of persistent system failures or performance bottlenecks

03Onboarding to a large, unfamiliar legacy codebase or complex architecture

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add rysweet/amplihack investigation-workflow

For use in Claude.ai and ChatGPT

Download Skill

Investigation Workflow for Claude Code

Investigation Workflow for Claude Code

Características Principales

Casos de Uso

Características Principales

Casos de Uso