kolla-create-architecture-markdown

Generate structured architecture Markdown from mapped codebase evidence including entrypoints, modules, workflows, configuration, and data flow.

Updated May 3, 2026
One-click install
npx skills add https://github.com/tabesink/deepdoc-agent --skill kolla-create-architecture-markdown
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kolla-create-architecture-markdown
Source: https://github.com/tabesink/deepdoc-agent/tree/main/.cursor/skills/kolla-create-architecture-markdown
Command: npx skills add https://github.com/tabesink/deepdoc-agent --skill kolla-create-architecture-markdown

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you create durable, evidence-backed architecture Markdown by first reviewing an existing codebase, so maintainers can understand how the system works without relying on guesswork.

Core Features & Use Cases

  • Codebase-informed architecture writing: Produces architecture documentation grounded in actual entrypoints, modules, configuration, and workflows discovered in the repository.
  • Structured, implementation-ready outputs: Writes a full system architecture document with a consistent section progression and concrete responsibilities, invariants, operations, and failure modes.
  • Practical guidance for agents and junior devs: Includes agent operating rules, quality gates, and a minimal first task so future coding agents can safely navigate and change the system.

Quick Start

Ask the AI to review your repository and generate an architecture document in the style of docs/architecture/llm-wiki-system-architecture.md for the feature or module you specify.

Frequently Asked Questions about kolla-create-architecture-markdown

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

FAQPage Schema
How do I generate architecture documentation from an existing codebase?

You can generate evidence-backed architecture documentation by mapping repository entrypoints, data flow, and modules to produce structured Markdown with operational outlines and explicit open questions.

What is the best way to write system design docs grounded in actual code?

The best way to write system design docs grounded in code is to map repository modules and data flow first, producing documentation that details concrete responsibilities, invariants, and failure modes.

Can I create implementation-ready architecture Markdown for developer onboarding?

Yes, you can create implementation-ready architecture Markdown for developer onboarding by generating documents that include agent operating rules, quality gates, and a minimal first task for safe system navigation.

How does an LLM agent workflow create architecture docs without guesswork?

An LLM agent workflow creates architecture docs without guesswork by mapping repository evidence like entrypoints and workflows first, then applying structured section patterns and evidence rules to scaffold the documentation.

Does this architecture documentation scaffolding work for reviewing specific product subsystems?

Yes, architecture documentation scaffolding works for reviewing specific product subsystems by applying its evidence rules and section progression across any specified repository, product area, or subsystem.