Analysis updated 2026-05-18
Scan a Go, JS/TS, or Python project and view architectural metrics in a local dashboard.
Compare a module's coupling, complexity, and size against the previous scan.
Export an architecture report as JSON or Markdown for documentation.
Block a CI pipeline merge automatically when a distance threshold is violated.
| mpraes/archi | aeneasr/github-trends | bxcodec/go-simple-flatbuffer | |
|---|---|---|---|
| Stars | 4 | 4 | 4 |
| Language | Go | Go | Go |
| Last pushed | — | 2020-12-18 | 2017-06-09 |
| Maintenance | — | Dormant | Dormant |
| Setup difficulty | easy | easy | easy |
| Complexity | — | 2/5 | 2/5 |
| Audience | developer | pm founder | developer |
Figures from each repo's GitHub metadata at analysis time.
Single static binary, AI enrichment is optional and needs a Gemini API key.
archi is a command line tool that scans a codebase and measures how well it is architected, then shows the results in a local dashboard that runs entirely on your own machine. It supports Go, JavaScript, TypeScript, and Python projects, including repositories that mix several of those languages, and it can either detect the language automatically or be told which one to use. It ships as a single binary with the web interface built in, so there is nothing extra to install once it is downloaded. Running the tool against a project opens a browser dashboard showing metrics for each module, along with how those numbers changed since the last scan. Clicking into a module reveals more detail on things like how tightly it is coupled to other parts of the code, whether it has become a bloated god module or an orphaned one nobody depends on, and how complex it is. If a source file has a syntax error, the tool warns about it and skips that file rather than stopping the whole scan. There is an optional AI enrichment step that adds extra insights on top of the computed metrics, using Google's Gemini model. This step only runs if an API key is supplied, and the dashboard's basic metrics load and display before the AI step even starts, so a slow or missing AI response never blocks the initial view. The README states that only the metric numbers are sent to the AI model, never the actual source code or the API key itself. Installation is available through several package managers depending on the operating system, including Homebrew, the Arch Linux AUR, the mise version manager, and direct DEB or RPM downloads, or the tool can be built from source using Go and npm. Beyond the interactive dashboard, archi also has headless commands meant for scripts and continuous integration pipelines: one exports a report as JSON or Markdown, and another checks the code against a distance threshold and exits with an error code if it is violated, which a team could use to block a merge automatically. The project is written in Go, with the web frontend embedded into the binary, and it is released under the MIT license. It is aimed at developers who want a quick, visual sense of a codebase's architectural health without manually digging through the code.
A local CLI tool that scans your codebase, scores its architecture, and shows metrics in a browser dashboard.
Mainly Go. The stack also includes Go, TypeScript, Python.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Verify against the repo before relying on details.