Investigation Workflow

This is a structured 6-phase workflow for investigating existing codebases when you need to actually understand how something works before changing it. It's normally invoked automatically by dev-orchestrator, but you can run it standalone via the recipe runner. The workflow goes from scope definition through parallel deep dives (multiple agents exploring simultaneously) to verification and knowledge capture. What's notable is the emphasis on not reading the markdown docs and following phases manually. The recipe runner enforces the actual phase ordering and agent deployment. Use this when you're asking "how does authentication work" or "why is CI failing," not when you're ready to implement something.

skills.sh installs: 249
GitHub stars: 76
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add rysweet/amplihack --skill investigation-workflow --agent claude-code

Installs into .claude/skills of the current project.

Files

Investigation Workflow Skill

Relationship to Dev Orchestrator

Normal execution path: This workflow is invoked as a sub-recipe by the dev-orchestrator skill via smart-orchestrator. You do NOT normally need to activate this skill directly.

User request → dev-orchestrator → smart-orchestrator recipe
    → investigation-workflow recipe (this skill's recipe)

Direct invocation is supported as a compatibility path when the dev-orchestrator is unavailable or when explicitly requested. In that case, use the recipe runner (see Execution Instructions below).

Workflow Graph

flowchart TD
    INIT[Initialize Tracking] --> P1

    subgraph P1["Phase 1: Scope Definition"]
        SCOPE[scope-definition<br/>prompt-writer agent] --> AMB{Has ambiguities?}
        AMB -->|yes| CLARIFY[clarify-ambiguities<br/>ambiguity agent]
        AMB -->|no| P1_OUT[Scope defined]
        CLARIFY --> P1_OUT
    end

SKILL.md on GitHub