skills.rest
BitRaptors
BitRaptorsOFFICIAL
Explore all skills by this author →
WebsiteGitHub

archie-deep-scan

Generate a semantic architecture blueprint with per-folder context files and health metrics.

16|Updated Jan 5, 2026
Search Tags:#architecture analysis#coding agents#codebase governance#incremental scanning#repository blueprint#architectural rules
One-click install
npx skills add https://github.com/BitRaptors/Archie --skill archie-deep-scan
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: archie-deep-scan
Source: https://github.com/BitRaptors/Archie/tree/main/archie/assets/workflow/deep-scan
Command: npx skills add https://github.com/BitRaptors/Archie --skill archie-deep-scan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents coding agents from losing architectural context, misplacing files, reimplementing existing helpers, and violating established design decisions as a repository evolves.

Core Features & Use Cases

  • Comprehensive Architecture Analysis: Runs a nine-step, two-wave scan covering structure, patterns, technology, data, domain laws, design decisions, risks, and architecture overview.
  • Persistent Agent Context: Produces blueprint files, per-folder CLAUDE.md and AGENTS.md context, architectural rules, dependency insights, findings, pitfalls, and health metrics.
  • Resumable and Incremental Workflows: Supports full, incremental, comprehensive, resumed, and step-specific scans with persisted state and validation.
  • Use Case: Run an initial baseline after joining a project, then use incremental scans after major changes so coding agents retain repository-specific architectural intent.

Quick Start

Use the Archie deep scan skill to analyze the current repository and generate its architecture blueprint, folder context files, rules, findings, and health metrics.

Frequently Asked Questions about archie-deep-scan

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

FAQPage Schema
How do I build an architecture blueprint so my coding agents understand my repository's design intent?▼

Building an architecture blueprint requires a nine-step, two-wave scan of your repository's structure, patterns, technology, data, and design decisions. The scan produces persistent context files, architectural rules, and health metrics that coding agents use to maintain repository-specific intent.

What is the best way to preserve architectural context and prevent coding agents from misplacing files in a large codebase?▼

Preserving architectural context for coding agents is achieved by generating per-folder CLAUDE.md and AGENTS.md files alongside a comprehensive repository blueprint. These outputs capture structural patterns, domain laws, and design decisions, preventing agents from reimplementing existing helpers or violating established rules.

Can I run an incremental scan to update my repository blueprint after a major refactor?▼

Incremental scans are fully supported to update your repository blueprint after major refactors. The analysis uses persisted state to resume interrupted scans and offers step-specific scans, ensuring coding agents retain updated architectural intent without requiring a full baseline rescan.

How does repository scanning and parallel AI reasoning work for generating architectural rules?▼

Repository scanning with parallel AI reasoning works by coordinating a two-wave analysis covering structure, patterns, data, domain laws, risks, and architecture overview. This parallel reasoning synthesizes raw findings into architectural rules, dependency insights, and per-folder context for coding agents.

Does this architecture analysis tool work without external dependencies or components?▼

This architecture analysis tool operates with no external dependencies or components listed. It runs independently through Archie tooling to coordinate repository scanning, parallel AI reasoning, rule synthesis, validation, telemetry, and health measurement for your codebase.

When should I use a comprehensive scan versus a step-specific scan for codebase governance?▼

Comprehensive scans are ideal for first-time baselines or major refactors requiring a full nine-step analysis. Step-specific scans are used when you need targeted updates to specific architectural layers, leveraging persisted state to avoid redundant processing across the repository blueprint.

Related Skills

Explore More Software Engineering Skills→
tachiiri-orgtachiiri-org

setup-role-python

Reconcile a local Python engine repository to its expected role baseline.

Official
Advanced
jlawsjlaws
2

code-agent-meta-patterns

Standardize CLAUDE.md structure and layering for code-agent workflows.

Community
Advanced
Luohao-YanLuohao-Yan
6

classify-changelog-pr

Classify pull requests into changelog categories using predefined rules.

Community
Intermediate
phuhao00phuhao00
12

code-analyst-agent

Analyze source code to identify symbols, call graphs, and architectural patterns.

Community
Intermediate