shrivatsas/dev-setup/architecture-docs

Claude Code and agent skill by shrivatsas/dev-setup

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Generate or update architecture documentation from repositories, product docs, infrastructure files, and code. Use when asked for high-level architecture, C4 diagrams, system context, container/component/deployment views, arc42 documentation, Excalidraw-ready architecture, Mermaid architecture diagrams, architecture decisions, runtime views, quality attributes, risks, or an architecture package synthesized from existing docs and source.

Install

npx vskill@latest install shrivatsas/dev-setup/architecture-docs

SKILL.md

The same SKILL.md was published earlier as shrivatsas/dev-setup/architecture-docs.

Architecture Docs

Use this skill to synthesize architecture documentation from a codebase and its docs. It supports:

  • C4: diagram-first context, container, component, and deployment views.
  • arc42: narrative architecture documentation with goals, constraints, decisions, runtime flows, quality requirements, risks, and glossary.
  • Combined package: C4 diagrams referenced from or embedded into arc42.

Workflow

  1. Discover source material
  • Search before reading deeply: use rg --files, then targeted rg for product names, architecture, infra, auth, integrations, deployment, SLOs, metrics, permissions, and feature docs.
  • Read primary local sources first: product README, workspace docs, backend/frontend READMEs, service docs, infra/Terraform, API routers, package files, and existing architecture docs.
  • For large repos, summarize by domain instead of reading every source file.
  1. Establish the boundary
  • Identify users/actors, in-scope system, adjacent products, external systems, data stores, identity providers, async workers, observability, and deployment platform.
  • Separate documented fact from inference. Explicitly mark uncertain deployment/runtime assumptions.
  1. Select output mode
  • If the user asks for diagrams, Excalidraw, C4, or high-level architecture: produce C4.
  • If the user asks for arc42, architecture docs, decisions, risks, or onboarding/reference material: produce arc42.
  • If the user asks for a complete architecture set/package: produce both and cross-reference them.
  1. Create artifacts
  • Prefer Markdown files in an appropriate docs/architecture location, unless the user names a path.
  • For C4, include Mermaid flowchart blocks that can be pasted into Excalidraw.
  • For arc42, reference C4 diagrams rather than duplicating every diagram inline unless the user requests a single self-contained file.
  1. Review before final
  • Check diagrams for basic Mermaid syntax hazards.
  • Check that claims are traceable to source docs or clearly labeled as assumptions.
  • Include unresolved questions or doc reconciliation items when sources disagree.

Reference Files

Read only the reference needed for the requested output:

  • references/c4.md: C4 output guidance and diagram checklist.
  • references/arc42.md: arc42 section structure and content checklist.
  • references/mermaid-excalidraw.md: Mermaid conventions for Excalidraw import.

Output Standards

  • Keep diagrams coarse enough for architecture discussion; avoid mirroring every class, table, endpoint, or file.
  • Use business/domain labels, then add technology in secondary text.
  • Include the operationally important things: auth, permissions, data stores, async work, external dependencies, observability, and deployment uncertainty.
  • For healthcare, financial, legal, or other sensitive domains, call out privacy, auditability, access control, data retention, and external data handling.
  • Do not treat generated architecture as authoritative if the repo docs conflict. Surface the conflict directly.

Suggested Filenames

  • C4: architecture-c4.md, <product>-c4-architecture.md, or <product>-c4-architecture-excalidraw.md
  • arc42: architecture-arc42.md or <product>-arc42-architecture.md
  • Combined index: architecture.md

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/shrivatsas/dev-setup/architecture-docs/badge)](https://verified-skill.com/skills/shrivatsas/dev-setup/architecture-docs)

Something wrong with this skill or its page? Tell us.