tt-a1i/archify
An AI agent skill that turns a described codebase into a verified, shareable architecture diagram inside chat
Archify is an agent skill that converts a codebase or system description into an interactive system map right in a chat session. It installs into AI coding tools like Cursor, Claude Code, Codex CLI, and OpenCode, and the output is a single self-contained HTML file that can be shared or exported. Before delivering a diagram, it runs the description through automated checks so the map doesn't invent connections that don't exist.
What it does
- Supports five diagram types—Architecture, Workflow, Sequence, Data Flow, and Lifecycle—plus dark/light themes and four visual presets.
- Turns a natural-language description into a typed JSON intermediate representation (IR), then requires it to pass schema, layout, HTML/SVG, and route checks before the final artifact replaces the last verified one.
- Offers an Architecture Delta feature comparing Before/Delta/After snapshots to show exactly what was added, removed, changed, or moved.
- Interactions like node search, upstream/downstream tracing, route probing, role comparison, and guided story playback all reuse the diagram's authored relationships instead of inventing new topology.
- Final output exports as PNG, SVG, WebM, and a 1200×630 share card, all bundled from one self-contained HTML file.
- Includes a real example: it traced the public repository mco-org/mco at commit 9f1a1cf and produced a working map from it.
Why it matters
Developers can get a verified architecture diagram from a short chat prompt instead of drawing it by hand, useful for documenting or explaining complex systems to a team. The before/after comparison feature also helps reviewers spot structural changes clearly during code review or pull requests.
Terms in this repo
- Agent Skill · An add-on capability installed into an AI coding assistant to perform a specific task
- JSON IR (intermediate representation) · A structured JSON format that stores diagram content, from which the final image is generated
- self-contained HTML · A single web file that holds all diagram visuals and behavior with no external dependencies
- upstream/downstream reach · A feature that traces the components connected before or after a chosen node
- share card · A 1200×630 image formatted for easy posting to documents or social media
Repository description (English)
Agent skill for beautiful, verifiable architecture, workflow, sequence, data-flow, and lifecycle diagrams—self-contained HTML with motion and crisp export.
Open on GitHubTrending repos
- bookorbit/bookorbitA self-hosted book server that syncs your reading progress across web, Kobo, and KOReader
- ConardLi/garden-skillsA skill pack that hands AI coding agents ready-made expertise for design, video, images, and search
- browser-use/browser-useA tool that lets AI click buttons and type into websites just like a human, to finish tasks for you
- K-Dense-AI/scientific-agent-skills163 how-to manuals that turn AI coding assistants into science research helpers
- DietrichGebert/ponytailA ruleset that teaches AI coding agents to say no to writing code
- liustack/modlensA plugin that lets text-only AI models like DeepSeek 'see' pasted images by turning them into structured text data
- marin-community/marinMarin is an open-source platform that publishes the entire process of building large language models, not just the final model
- TauricResearch/TradingAgentsA framework where multiple LLM agents split into analyst, researcher, trader and risk-management roles, debate, and decide stock trades
Latest from METAL LAB
- Ox Alpha turns out to be GLM-5.3-Flash
- Martin Casado says AI has erased startups' distribution barrier
- OpenAI introduces Runme, which uses Codex to document repetitive work
- Stable Diffusion Maker Raises $76M From Music, Game Studios
- Meta's paid agent Hatch to launch alongside new Watermelon model in October