manage-architecture

Merge derived module data with LLM-enriched annotations to map project architecture.

4|Updated Jan 11, 2026
One-click install
npx skills add https://github.com/cuioss/plan-marshall --skill manage-architecture-cuioss
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: manage-architecture
Source: https://github.com/cuioss/plan-marshall/tree/main/marketplace/bundles/plan-marshall/skills/manage-architecture
Command: npx skills add https://github.com/cuioss/plan-marshall --skill manage-architecture-cuioss

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Automates the transformation of raw project data into a structured architectural view for planning and execution.

Core Features & Use Cases

  • Discovery: Collect raw module data from extensions and build a coherent module map.
  • Enrichment: LLm analyzes documentation and code to assign responsibility, purpose, dependencies, and skills.
  • Persistence: Persist enriched data to llm-enriched.json for downstream consumption and queries.
  • Use Case: For a multi-repo project, automatically generate a solution-outline-ready representation for module placement decisions.

Quick Start

Execute the documented workflow to perform discovery, enrichment, and persistence for your project architecture.

Frequently Asked Questions about manage-architecture

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

FAQPage Schema
How do I map and enrich project architecture for multi-module planning?

Mapping project architecture involves merging derived module data with LLM-enriched annotations. This coordinates discovery and persistence across multi-module projects to produce a structured representation for downstream planning.

What is LLM enrichment for project module responsibility and dependencies?

LLM enrichment analyzes documentation and code to assign module responsibility, purpose, dependencies, and skills. It transforms raw project data into a structured architectural view and persists enriched data to llm-enriched.json for downstream consumption.

Do I need local data exports to generate a structured architectural view?

Yes, you need local data exports under .plan/project-architecture. Specifically, derived-data.json and llm-enriched.json are required to execute the documented workflow for discovering and enriching your project architecture.

Can I use this architecture discovery workflow for multi-repo projects?

Yes, architecture discovery supports multi-repo and multi-module projects. It collects raw module data from extensions to automatically generate a solution-outline-ready representation for module placement decisions.

What's the best way to automate transformation of raw project data for execution planning?

Automating raw project data transformation requires executing a documented workflow of discovery, enrichment, and persistence. This merges derived module data with LLM annotations to persist a structured architectural view for planning and execution.

Why does the architecture enrichment workflow require strict script execution?

Architecture enrichment requires strict script execution to coordinate discovery and persistence across multiple modules accurately. It enforces the documented workflow to ensure derived data and LLM annotations merge correctly into the structured representation.