pathfinder

Map codebases into feature-grouped Mermaid flowcharts with duplication reports.

1|Updated May 16, 2026
One-click install
npx skills add https://github.com/enigmaicon-eng/AI-Enterprise-OS --skill pathfinder-enigmaicon-eng
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pathfinder
Source: https://github.com/enigmaicon-eng/AI-Enterprise-OS/tree/main/examples/claude-mem/plugin/skills/pathfinder
Command: npx skills add https://github.com/enigmaicon-eng/AI-Enterprise-OS --skill pathfinder-enigmaicon-eng

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you map an unfamiliar or sprawling codebase into clear feature boundaries, identify duplicated concerns with citations, and propose the simplest unified architecture before refactoring.

Core Features & Use Cases

  • Feature inventory with boundaries: Produces a structured feature list based on source-tree evidence and entry points.
  • Per-feature flowcharts: Generates Mermaid flowcharts (with file:line node labels) for each feature’s primary happy path and key branches.
  • Duplication and unification proposals: Produces a cross-feature duplication report and a unified architecture proposal, then generates copy-paste /make-plan handoff prompts.

Quick Start

Ask the AI to run pathfinder on the repository to find the ideal refactor path and produce unified architecture plus per-system /make-plan prompts.

Frequently Asked Questions about pathfinder

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

FAQPage Schema
How do I map an unfamiliar codebase into feature boundaries for refactoring?

To map an unfamiliar codebase for refactoring, you need to inventory features based on source-tree evidence and entry points. This process generates structured feature lists with clear boundaries and per-feature Mermaid flowcharts detailing primary happy paths.

What is the best way to detect duplicated concerns across multiple interacting systems?

Detecting duplicated concerns across systems requires a subagent-based discovery approach with mandatory source citations. This method produces a cross-feature duplication report identifying overlapping logic with specific file and line evidence.

How do I generate Mermaid flowcharts for code architecture audits?

Generating Mermaid flowcharts for architecture audits involves mapping each feature's primary happy path and key branches. The resulting diagrams include specific file:line node labels, providing visual evidence of codebase structure and flow.

Can I use this approach to plan a unified architecture across multiple systems?

Yes, you can plan a unified architecture across multiple systems. After detecting duplicated concerns, the process proposes the simplest unified architecture supported by evidence, fitting architecture audits and refactor preparation tasks.

How do I hand off a refactoring plan to another agent or tool?

To hand off a refactoring plan, you generate copy-paste /make-plan prompts. These handoff prompts are produced alongside the unified architecture proposal, ensuring downstream execution tools receive structured, evidence-based instructions.

When should I not use a subagent-based discovery approach for codebase mapping?

You should avoid subagent-based codebase mapping when working with extremely small or simple codebases where manual inspection suffices. This approach is designed for sprawling, unfamiliar codebases requiring cross-feature duplication detection and unified architecture proposals.