RepoDNA v1.3.1: Analyze a Codebase in Your Browser—Without Uploading It
Understanding an unfamiliar codebase often means answering the same questions from scratch:
- How is the project organized?
- Which modules and packages depend on one another?
- What has changed over time?
- Which files are hotspots?
- How is the project built and tested?
- Which findings have evidence behind them? RepoDNA is an open-source, local-first repository intelligence and code archaeology tool designed around those questions. It builds architecture maps, histories, quality signals, reports, comparisons, onboarding guides, and shareable Project DNA cards. Findings link back to files, lines, commits, or measurements so developers can inspect the evidence. Today, RepoDNA v1.3.1 adds a new way to get started: analysis directly in the browser. Analyze a folder or archive in the browser Open the RepoDNA web version, choose a folder or a supported archive, and let the WebAssembly build analyze it in the browser. You don’t need to install the native CLI just to try the core analysis, and the project files aren’t uploaded. The browser interface can also create HTML and Markdown reports and Project DNA cards that can be downloaded as SVG or PNG. After the web version has loaded, it can continue working offline. One important boundary: the browser cannot read a repository’s Git history, so this mode does not include historical analysis. Use the command line or desktop app when you need commit timelines, contributor history, file evolution, or the Codebase Time Machine. What else is in v1.3.1?
- WebAssembly package: @sanskarin/repodna-wasm brings RepoDNA analysis, reports, and cards to Node.js and browsers. For example, npx @sanskarin/repodna-wasm analyze . can analyze a folder or archive.
- GitHub Action: use sanskarIN/RepoDNA@v1.3.1 in a workflow to show a summary, annotate findings on files, keep the analysis, optionally write a full report, compare with a baseline, and choose the severity at which a step should fail. The Action runs on the runner; it does not upload the checked-out source to a RepoDNA service.
- Windows on Arm: the CLI and desktop app now have Windows Arm builds.
- Desktop saving fixes: full-report folders are kept separate by analysis, and Project DNA cards can be saved in the theme shown by the preview. The Action is listed in the GitHub Marketplace. Its documentation includes the inputs, outputs, and workflow examples. Try it
- Browser demo: https://sanskarin.github.io/RepoDNA/
- Repository and v1.3.1 release: https://github.com/sanskarIN/RepoDNA/releases/tag/v1.3.1
- GitHub Marketplace Action: https://github.com/marketplace/actions/repodna
- Changelog: https://github.com/sanskarIN/RepoDNA/blob/main/CHANGELOG.md RepoDNA is evidence-first and descriptive, not a grade for a project or its contributors. The analysis is lexical rather than compiler-level, and the tool labels analysis depth and reports missing data rather than treating it as zero. I’d welcome feedback on what makes a repository easier to understand and which views are most useful in day-to-day maintenance. A question for developers When you inherit an unfamiliar project, what do you inspect first: architecture, dependency relationships, Git history, hotspots, or tests/build setup? I’m especially interested in feedback from people maintaining multi-language repositories or projects with a long history.
Top comments (0)