technical-discovery

# Technical Discovery

Perform technical research and analysis to understand a problem, question, or proposed change. Produce a comprehensive report without any code implementation.

## Goal

Investigate the codebase and relevant context to provide findings that inform decision-making or subsequent issue creation. This skill answers "what would be involved?" without actually doing the work.

## Required Inputs

- **Problem or Question**: Description of what to analyze (e.g., "How would we add user authentication?" or "Why is the API slow?")
- **Additional Context** (optional): Constraints, related issues, or specific areas to focus on

## When to Ask Questions

Use `AskUserQuestion` to clarify before proceeding when:

- The problem statement is too vague to research meaningfully
- Multiple interpretations exist and the correct one significantly affects the analysis
- You discover a decision point where user input would focus the research
- You need domain knowledge or context that isn't in the codebase

Ask focused questions early rather than making assumptions. One good clarifying question up front saves wasted research down the wrong path.

## Process

### 1. Understand the Request

Parse the problem or question to identify:

- **Core question**: What specifically needs to be answered?
- **Scope boundaries**: What's in scope vs. out of scope for this analysis?
- **Success criteria**: What would a useful answer look like?

### 2. Research the Codebase

Investigate relevant areas:

- **Find affected files**: Use Glob and Grep to locate code related to the problem
- **Read key files**: Understand current implementations, patterns, and conventions
- **Trace dependencies**: Follow imports and call chains to understand relationships
- **Check tests**: Review existing tests to understand expected behavior

### 3. Research External Context

When the problem involves libraries, APIs, or concepts outside the codebase:

- **Use Perplexity**: Query for up-to-date documentatio