Engineering
wiki-architect avatar

wiki-architect

Analyzes codebases to generate hierarchical documentation, onboarding guides, and architectural mapping, helping teams understand and document their projects efficiently.

Introduction

The wiki-architect skill is a sophisticated documentation tool designed for software engineers, technical leads, and project managers. It automates the extraction of architectural insights from local or remote repositories, transforming raw codebase information into a structured wiki format. This agent performs deep analysis of project structures, framework usage, and domain models to create onboarding paths for diverse audiences, including new contributors, staff engineers, executive leadership, and product managers. By enforcing a rigorous resolution process for source code context, it ensures that every generated section includes accurate citations to specific file paths or repository lines, maintaining traceability.

  • Automatically maps project architecture, including layers like presentation, business logic, data access, and infrastructure.

  • Generates tailored onboarding documentation: contributor guides with language-specific foundations, staff engineer guides focusing on design tradeoffs and decision logs, and executive/PM guides that distill technical information into business value.

  • Supports multi-language project analysis, utilizing specific comparison languages (e.g., Python to JS, Rust to Go) to explain technical concepts.

  • Employs strict hierarchical structuring with support for table-of-contents generation, onboarding flows, and deep-dive technical explorations.

  • Integrates directly into agent workflows by verifying git remotes and branch contexts before execution to ensure accuracy.

  • Ideal for quickly generating project wikis, updating onboarding materials, or performing architectural reviews during handoffs.

  • Requires resolving source repository context (git remote or local paths) as a prerequisite to prevent hallucination.

  • Designed for high-density information output, with constraints on nesting depth and node counts to ensure readability and token efficiency.

  • Encourages deep-linking by using URL-based citations for remote repositories or line-number references for local files.

  • Best used when the user requires a comprehensive zero-to-hero project overview or a structured documentation plan for complex codebases.

Repository Stats

Stars
2,174
Forks
249
Open Issues
48
Language
TypeScript
Default Branch
main
Sync Status
Idle
Last Synced
Apr 29, 2026, 12:29 PM
View on GitHub