DevSecOps engineer sharing experiences and insights on AI, Web3, and Claude code development. Aims to help solve technical challenges, improve development efficiency, and grow through community interaction. Feel free to comment and discuss.
Claude Code 1.0.123 upgrades user-defined slash commands into AI‑driven tools, allowing the Claude assistant to autonomously invoke commands like /lint or /deploy‑staging, turning manual shortcuts into automated actions that streamline development workflows.
This guide explains how to configure Jest test scripts in a package.json file and use npm commands to run the full test suite, generate coverage reports, watch for changes, and target individual test files or cases for faster, more focused development.
This guide explains how to use Git's modern "git restore" command, the classic "git checkout" syntax, and "git clean" to safely discard tracked changes or remove untracked files in a chosen directory, including safety tips like dry‑run options.
This article reviews essential tools and formats—such as Postman, Apidog, Swagger/OpenAPI, and API gateways like Amazon API Gateway and Kong—that help backend developers efficiently save, document, and manage HTTP API information throughout the development lifecycle.
This guide explains why SSH agent forwarding can break despite correct client and server settings, reveals the hidden role of connection multiplexing, and provides a step‑by‑step solution to reset the master connection so forwarding works again.
This guide explains how macOS Keychain can store and automatically provide SSH key passphrases, combining strong encryption with seamless daily use by configuring the SSH client, adding the key to the agent, and updating the ~/.ssh/config file.
This article examines script injection threats in Java applications and compares static, dynamic, and interactive security testing tools—SAST, DAST, and IAST—highlighting their strengths, weaknesses, and popular options to help developers select the most suitable solution.
This article examines the design philosophy behind the “Bash Only” category of the SWE‑bench benchmark, explaining how its minimal‑agent approach isolates LLM reasoning by restricting interactions to a plain Bash shell, making it a rigorous, reproducible test of true software‑engineering intelligence.
The article compares two AI‑assisted development workflows—the traditional “question‑answer consultant” model and the emerging paid “coding teammate” approach—explaining how context window size, model call frequency, tool integration, and stateful collaboration make free plans insufficient for serious developers.
The article analyzes the architectural and design challenges introduced by AI programming assistants, introduces the open‑source spec‑driven framework spec‑kit as a structured solution, and explains how it reshapes developer roles while offering a strategic advantage in the emerging AI‑augmented software era.
