Claude Code has released version 2.1.157, bringing significant improvements to plugin management and user experience. This update allows automatic loading of plugins from .claude/skills directories, removing the need for a marketplace. It also introduces autocomplete for plugin arguments and respects theagent field in settings.json for dispatched sessions. The update addresses several bugs, including issues with unprocessable images and terminal freezing, enhancing overall stability. These enhancements make Claude Code more efficient and accessible for developers.
Read originalClaude Code's latest update, v2.1.158, introduces Auto mode for Bedrock, Vertex, and Foundry platforms, specifically for Opus 4.7 and 4.8. This feature can be activated by setting the environment variable CLAUDE_CODE_ENABLE_AUTO_MODE to 1. The addition of Auto mode aims to streamline processes and enhance user experience across these platforms. While the update doesn't introduce groundbreaking changes, it offers a practical enhancement for developers working with these specific versions.
The latest update to Claude Code, version 2.1.153, introduces several enhancements and fixes that improve user experience and functionality. Notably, the update adds a skipLfs option to the GitHub plugin marketplace, allowing users to bypass Git LFS downloads, and enhances the autocomplete feature for Claude agents. Additionally, it addresses various bugs, including memory usage issues and installation errors on Windows. These changes streamline operations and ensure smoother interactions with the platform, particularly for developers using Claude Code in complex environments.
The latest update to Claude Code, version 2.1.154, introduces several enhancements aimed at improving task management and efficiency. Notably, the new dynamic workflows feature allows users to orchestrate complex tasks across multiple agents, making it easier to handle larger projects. The update also brings cost-effective fast mode on Opus 4.8, offering increased speed at a reduced rate. Additionally, the release refines the system's decision-making process by reserving multiple-choice prompts for genuinely uncertain situations. These changes collectively enhance the platform's usability and performance, particularly for developers managing intricate workflows.
© GitHub ChangelogGitHub has enhanced its Copilot usage metrics API by introducing AI adoption phases, allowing enterprises to better understand how users engage with Copilot. This new feature classifies users into four phases based on their interaction with Copilot over a 28-day period, providing insights into their adoption journey. The API now includes a field for ai_adoption_phase, offering detailed metrics at both user and organizational levels. This development enables enterprises to tailor training and support efforts, focusing on phases with the most potential for growth. By moving beyond simple active-user counts, organizations can now track user progression and optimize Copilot adoption strategies.
© Lev SelectorCursor has launched its Composer 2.5 model, aimed at improving coding efficiency.
Braintrust engineers are now using Codex, integrated with GPT-5.5, to enhance their coding efficiency and experiment execution. This integration allows them to swiftly convert customer requests into functional code, significantly reducing manual coding time. By adopting Codex, Braintrust can focus more on complex problem-solving rather than routine coding tasks. This approach exemplifies the increasing adoption of AI-assisted coding, which is set to boost productivity and drive innovation in software development. The shift towards AI tools in coding is reshaping how engineers approach their work, offering new possibilities for efficiency and creativity.