caru-ini/stack-diagram
Developer ToolsAn agent skill that turns a one-sentence stack description into an architecture diagram with official logos.
What it does
stack-diagram is a skill for coding agents: you describe your stack in a sentence (for example, Next.js and a Hono API on Vercel with Neon Postgres, Clerk auth and Stripe payments), and the agent produces an architecture diagram. It is aimed at developers who need a presentable system diagram without manually placing boxes and pulling logo files.
The skill handles logo sourcing from svgl, devicon, gilbarbara/logos, brand repositories, AWS's own icon package and Lucide, and it routes connector lines around labels. Diagrams are rendered in light and dark themes. Output can be a self-contained HTML file with fonts and logos inlined, a PNG, a dependency-free React component, or a Claude artifact.
Quick start
Install the skill with the skills CLI:
npx skills add caru-ini/stack-diagram
The README lists Claude Code, Codex, Cursor and other agents that support skills as compatible. In the Claude apps, upload stack-diagram.zip from the latest release as a skill instead. Then prompt your agent with a stack description, such as:
Diagram our stack: Next.js and a Hono API on Vercel. Postgres on Neon, auth with Clerk, payments with Stripe. GitHub Actions deploys.
The agent renders the diagram and checks it before handing it over. You can request additional output formats; the self-contained HTML file is the default.
Requirements and limits
- Python 3.9+ and network access for fetching logos are required.
- PNG export uses Playwright with Chromium.
- Logos are downloaded at build time rather than bundled with the skill, so logo availability depends on the listed sources and brand repositories. All trademarks belong to their owners, and using a logo in a diagram does not imply endorsement.
- Code and docs are MIT-licensed; bundled fonts are under the SIL Open Font License 1.1.
HotRepo's take
The project entered HotRepo's new-project list on 2026-11-01 with 132 stars in its first observation window, and no GitHub daily rank has been recorded. That single early snapshot only establishes initial traction, not a trend. The notable technical point is the skill packaging itself: it fits the growing pattern of distributing agent capabilities through the skills CLI and agentskills-compatible clients, while keeping rendering and logo resolution deterministic in a Python toolchain.
Written by AI from the project's README, updated . The project's own docs are the source of truth.