refresh-architecture

Regenerate architecture analysis artifacts for Python, TypeScript, and SQL migrations.

4|1|Updated Jan 16, 2026
One-click install
npx skills add https://github.com/jankneumann/agentic-coding-tools --skill refresh-architecture
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: refresh-architecture
Source: https://github.com/jankneumann/agentic-coding-tools/tree/main/.agents/skills/refresh-architecture
Command: npx skills add https://github.com/jankneumann/agentic-coding-tools --skill refresh-architecture

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires ts-morph, typescript, ts-node, psycopg2-binary, tree_sitter, and includes scripts (resource) components.

What problem does it solve?

Outdated or missing architecture-analysis artifacts make planning, impact assessment, and parallel work unsafe and slow by depriving engineers and agents of a canonical graph, flow traces, and visual views of the codebase. This Skill provides a reproducible, auditable pipeline to rebuild, validate, diff, and visualize architecture artifacts so teams and automation can make informed decisions.

Core Features & Use Cases

  • Regenerate a multi-layer architecture analysis pipeline that produces per-language analysis, a canonical architecture graph, cross-layer flows, high-impact rankings, and summary reports.
  • Validate existing graphs to detect broken cross-layer flows and schema issues without regenerating state, and run diffs against a baseline commit to assess architectural impact of changes.
  • Produce Mermaid views, parallel modification zone analysis, feature-slice diagrams, and CI-friendly diagnostics; useful before planning work, when validating a PR, or when triaging architectural regressions.

Quick Start

Run a full architecture refresh pipeline to regenerate docs/architecture-analysis artifacts including graphs, views, diagnostics, and summary reports.

Frequently Asked Questions about refresh-architecture

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

FAQPage Schema
How do I regenerate architecture analysis artifacts for a repository?

Regenerate architecture analysis artifacts by running a full pipeline refresh that analyzes Python, TypeScript, and SQL migrations to produce canonical graphs, cross-layer flows, and summary reports. The pipeline writes outputs directly to the docs/architecture-analysis directory.

How do I validate architecture graphs and detect broken cross-layer flows in CI?

Validate architecture graphs in CI by running validation hooks that detect broken cross-layer flows and schema issues without regenerating state. This CI-friendly diagnostic process checks existing artifacts to triage architectural regressions before merging changes.

Can I generate Mermaid diagrams for feature-slice analysis from TypeScript and Python?

Generate Mermaid diagrams for feature-slice analysis from TypeScript and Python repositories by running the views generation pipeline. This process produces visual views, parallel modification zone analysis, and feature-slice diagrams for architectural planning.

How do I diff architecture graphs against a baseline commit to assess impact?

Diff architecture graphs against a baseline commit to assess the architectural impact of changes by running the diffing process. This compares current canonical graphs with the baseline state to identify structural regressions and flow modifications.

Do I need Python 3.12 and npm to run the architecture analysis pipeline?

You need Python 3.12+ and npm to run the architecture analysis pipeline, as these are required for TypeScript analysis and the underlying scripts. Write access to the docs/architecture-analysis directory is also necessary to output the regenerated artifacts.

What problems does outdated architecture graph state cause for engineering teams?

Outdated architecture graph state makes planning, impact assessment, and parallel work unsafe by depriving engineers of canonical graphs and flow traces. Reproducible artifact regeneration solves this by providing auditable, visual views of the codebase for informed decisions.