SShortSingh.
TopTechnologyAI & MLStartupsProgrammingScienceBusinessFinanceCrypto & Web3World
Back to feed
Programming

The Last Maintainer: Open-Source Local AI Tool to Recover Knowledge From Legacy Code

0
·3 views

A developer has released The Last Maintainer, an open-source, local-first AI project aimed at helping engineers understand poorly documented legacy codebases. The tool plans to analyze source code, API endpoints, database entities, Git history, and documentation to reconstruct how an application works. Its design extracts verifiable facts through deterministic code analysis before using a local open-weight model via Ollama to explain them, with references developers can inspect. The initial focus is Java and Spring Boot applications, with other languages possibly supported later. Keeping inference local lets teams avoid sending sensitive code and configuration to external AI services.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Log in to join the discussion and vote.

Log in

Related stories

0
ProgrammingDEV Community ·

CSSForge: Free Tool Generates Layered CSS Shadows and Glassmorphism Code

A developer has released CSSForge, a free, lightweight web tool for creating multi-layered CSS shadows and glassmorphism effects. The article argues that single box-shadows often look harsh, while layered shadows add more depth and realism to UI components. The tool offers real-time previews in dark, light, gradient, and image modes, and exports code as pure CSS, CSS Variables, or Tailwind CSS. It runs entirely on the client side and requires no registration. The developer is also seeking user feedback on interface improvements and additional generators.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Kuyo: a Japanese-style memorial tool that lays deleted code to rest with local AI epitaphs

Kuyo is a tool that treats deleted source files as the dead, drawing on the Japanese kuyo tradition of memorializing objects that served well. It scans a git repository for removed files, following renames, and records each file's lifespan, time of death, commit size, test coverage, and final commit message. A local Gemma model running on Ollama writes an epitaph for each file, and the results are rendered as a self-contained HTML cemetery. The author demonstrated it on a private repository with 137 deleted files, where the code itself never left the machine. The project offers commands for scanning, burying, consulting a file's ghost, and running as a GitHub Action.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Apple Unveils 14.2-Inch Touch-Enabled OLED MacBook Pro With Pencil Support

At an October 28, 2026 keynote, Apple introduced a 14.2-inch MacBook Pro with a touch-enabled OLED display, marking a departure from its long-standing no-touch policy. The screen uses a capacitive sensor grid that supports multi-finger gestures and the second-generation Apple Pencil, including pressure and tilt detection. According to the article, the OLED panel offers sub-millisecond response times, 1,200 nits peak brightness, and full DCI-P3 color coverage. Existing macOS apps generally receive touch input automatically, though some may need an update through Apple's Compatibility SDK, and battery life is estimated at about 11 hours under typical use.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Mutation testing with mutmut and StrykerJS to check whether tests actually verify behavior

A developer noticed that a pull request with 94% coverage and passing CI still let code changes slip through, since altering a comparison operator or replacing a return value left the tests green. Coverage only shows which lines run, not whether tests check the results. The article introduces mutation testing, where tools generate small faulty versions of code called mutants and check whether the test suite catches them. It demonstrates the approach in Python with mutmut on a shipping fee function, showing how weak assertions and missing boundary cases let mutants survive. The author also outlines StrykerJS configuration for JavaScript and TypeScript projects, highlighting per-test coverage analysis and incremental runs to keep execution fast.

0 comments Read more at DEV Community