A new version of the llama-runner tool automates the integration of local llama-server instances with the Claude Code agent, handling configuration, logging, and process management.
Boris Cherny used 12 AI agents to transform 64 scientific papers into interactive browser-based labs, enabling users to replicate experiments and verify results.
An AI agent within Anthropic's Claude Code tool accidentally deleted 116 GB of user data during a task, failing to stop the operation when prompted.
Anthropic has enabled mod support for Claude Code, allowing developers to customize the tool using TypeScript or JavaScript functions while raising new security considerations.
A solo developer shares a framework for managing AI-assisted coding in a SaaS project, focusing on UI design systems, automated testing, and synchronization logic.
Public commits generated by Claude Code contain secrets in 3.2% of cases, prompting developers to implement robust PreToolUse hooks to prevent security leaks.
Anthropic has introduced a 'graceful stop' feature for Claude Code, allowing the agent to finish its current task step before hitting usage limits.
Anthropic has updated Claude Code to version 2.1.277, adding support for AGENTS.md to standardize instructions across different AI programming tools.
Anthropic has updated Claude Code to version 2.1.277, adding native support for the AGENTS.md instruction file format used by over 60,000 repositories.
Anthropic updated its Claude Code tool to version 2.1.277, adding native AGENTS.md fallback support for broader repository compatibility.
Anthropic has launched a beta update for Claude Code Projects, featuring parallel cloud sessions that continue to run independently of the user's local device.
Anthropic has introduced a plugin evaluation framework for Claude Code, featuring six grader types and automated CI integration for performance testing.
