Developer Builds Open-Source Bot to Automate GitHub Bug Reproduction

·1 views

A solo developer has released Ghost Hunter, an open-source CLI and webhook bot that automates bug reproduction for GitHub issues. The tool uses an LLM to parse issue text, spins up an isolated Docker container, runs the reproduction steps, and posts crash logs back to the issue thread. Shortly after launch, community members flagged two security vulnerabilities: prompt injection risks from untrusted issue text and insufficient network isolation within Docker containers. The developer acknowledged both concerns and shipped fixes, noting that Metabase's similar internal tool requires a human-in-the-loop trigger for the same reasons. Ghost Hunter is MIT-licensed and available on GitHub, though the developer warns that GitHub App authentication and free-tier LLM reliability remain areas of ongoing improvement.

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)

DevUtils MCP Server Brings 36 Developer Tools to Any Local AI Assistant

DevUtils MCP Server v1.1.0 is an open-source, MIT-licensed package that connects 36 common developer utilities — including JSON validation, JWT inspection, and CIDR calculation — to any MCP-compatible AI assistant. The server runs entirely locally over standard input/output, requiring no API key, account, or external service, only Node.js 18 or newer and npm. It can be added to clients such as Claude Desktop, Cursor, VS Code, and Windsurf by inserting a short configuration entry and restarting the client. Once configured, the AI assistant calls explicit tools rather than attempting to perform operations from memory, reducing interpretation errors. The package is available on npm and can be run without a global installation using npx.

0 comments Read more at DEV Community

ProgrammingHacker News ·

Small Modular Reactors Edge Nuclear Power Toward Mainstream Energy Use

Small modular reactors (SMRs) are emerging as a renewed path forward for nuclear energy, offering a more practical and cost-effective alternative to traditional large-scale plants. Unlike conventional reactors, SMRs are designed to be factory-built and deployed in a fraction of the time, reducing construction risks and upfront costs. Their compact size makes them suitable for a wider range of locations, including remote areas and industrial sites. Proponents argue that SMRs could help fill gaps left by intermittent renewable energy sources as nations seek to decarbonize their power grids. Several projects are currently in development or early deployment stages across multiple countries, signaling growing interest from both governments and private investors.

0 comments Read more at Hacker News

ProgrammingDEV Community ·

Developer Runs Local AI Models on AMD RX 9060 XT via WSL and ROCm on Windows

A developer and gamer successfully set up a local AI environment on a personal Windows desktop using an AMD Radeon RX 9060 XT GPU with 16 GB VRAM. The setup involved running Ubuntu 24.04 under Windows Subsystem for Linux (WSL), then installing Ollama along with its dedicated ROCm backend to enable GPU-accelerated inference. A key step was installing librocdxg version 1.2.2, which resolved a GPU detection issue present in the earlier 1.2.0 release, and setting the HSA_ENABLE_DXG_DETECTION environment variable to expose the GPU through the DXG interface. Ollama was configured as a systemd service within WSL, allowing it to run persistently and optionally be accessed from outside the WSL environment. The walkthrough demonstrates that AMD gaming hardware can be repurposed to run large language models locally without requiring dedicated server infrastructure.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Ota v1.6.26 validates generated artifact lineage across EventCatalog sibling packages

Software tool Ota was pressure-tested against the EventCatalog monorepo to verify that generated source artifacts are properly tracked from producer to downstream consumer. The test exposed a common monorepo blind spot where generated code may exist and its producer task may pass, yet the actual consumer package still fails due to unresolved dependencies. Ota models generated outputs as distinct artifacts with declared producers, source inputs, and consumer requirements, making lineage visible before and during execution rather than only at build failure. A CI matrix running on Ubuntu, macOS, and Windows executed the full VS Code extension consumer closure and archived execution receipts as evidence, with a separate Linux job generating a formal workflow proof. The verified scope is limited to one language-server-to-extension build path and does not cover the broader EventCatalog workspace, semantic correctness, or distribution behavior.

0 comments Read more at DEV Community