ix-investigate

Investigates a code symbol by building a graph-based model of its role and connections.

7|2|Updated Mar 28, 2026
One-click install
npx skills add https://github.com/ix-infrastructure/ix-claude-plugin --skill ix-investigate
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ix-investigate
Source: https://github.com/ix-infrastructure/ix-claude-plugin/tree/main/skills/ix-investigate
Command: npx skills add https://github.com/ix-infrastructure/ix-claude-plugin --skill ix-investigate

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you quickly understand what a specific symbol (function, feature, or bug candidate) is and how it fits into your codebase, without getting lost in large files or excessive searching.

Core Features & Use Cases

  • Graph-first investigation: Locates and explains the target using Ix graph data before considering any code reads.
  • Evidence-driven execution flow: Produces connections (callers/callees) and a trace when needed, then stops once the execution path is clear.
  • Minimal and bounded source reading: Falls back to a single targeted read only if earlier phases can’t resolve the remaining question.
  • Optional Pro design context: Can include relevant recorded decisions affecting the symbol when Ix Pro is available.
  • Optional saving of results: Supports a save mode that writes the final investigation output to a chosen (or auto-generated) markdown path.

Quick Start

Ask: ix-investigate --save bug-foo.md "how does FooHandler work?" to generate a bounded, graph-based investigation and save the report.

Frequently Asked Questions about ix-investigate

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I investigate a function symbol and understand its execution flow without reading excessive source code?

You can investigate a function symbol by building a graph-backed mental model of its role and execution flow. This approach uses graph data to locate and explain the target, producing caller and callee connections to clarify the execution path before any source reading.

What's the best way to perform a root cause analysis on a bug candidate in a large codebase?

The best way to perform root cause analysis on a bug candidate is using graph reasoning to trace the symbol's upstream and downstream connections. This evidence-driven method maps the execution path and limits source scanning to at most one targeted code read if the graph cannot resolve the question.

When do I need graph-based code understanding instead of manual codebase exploration?

You need graph-based code understanding when you must deep-dive a specific feature or symbol without getting lost in large files. It locates and explains the target using graph evidence first, preventing excessive searching and unbounded scanning across the codebase.

Can I save the investigation report of a codebase deep-dive directly to a markdown file?

Yes, you can save the investigation report to a markdown file. By using the save mode with a chosen or auto-generated markdown path, the final graph-based execution flow and symbol analysis output is written directly to that file.

Does graph reasoning for call graph tracing support accessing recorded architectural decisions?

Yes, graph reasoning for call graph tracing can include relevant recorded decisions affecting the symbol when Ix Pro is available. This provides additional design context alongside the execution flow and caller or callee connections during the investigation.

What are the limitations of using a call graph for symbol investigation?

The main limitation is that graph data may not always fully resolve the execution path. When this happens, the investigation falls back to a single targeted source code read, bounding the effort but indicating the graph lacked sufficient evidence to complete the understanding.