AI Briefing#claude-code
The latest AI and developer news about #claude-code, with the original source and a short summary.
The latest AI and developer news about #claude-code, with the original source and a short summary.
A Claude Code plugin has been released that selectively deletes unnecessary tool call results without summarizing the conversation, utilizing TypeSafe AI's Jev model.
Worktrunk is a newly released CLI tool that automates Git worktree management for parallel AI coding agents such as Claude Code.
Released a coordinator-worker based 'Chief of Staff' pattern and operational loop to address context loss and lack of verification in Claude Code.
A six-month development of a Bach music generation system using Claude Code revealed that subscription fees accounted for only 2% of total costs, while the remaining 98% was spent on human judgment and verification.
The latest Claude Code update introduces memory usage warnings, a new environment variable for controlling MCP server connection wait times, and various bug fixes and stability improvements.
An effort_cost_index, representing relative token usage estimates for Effort levels by model, was discovered in the Claude Code binary.
A confirmed issue in Claude Code Auto Mode prevents nested CLAUDE.md files from loading due to Bash-first instructions.
A zero-click remote code execution vulnerability via SHA pinning bypass has been discovered in major AI coding agents including Claude Code and Codex, with patches currently in progress.
An optimization method has been released to reduce context rewrite costs after long tool calls by changing the default 5-minute cache TTL for Claude Code sub-agents to 1 hour.
JPMorgan has introduced an AWS container-based 'Devspace' environment for some engineers' use of Claude Code, blocking standing access to internal systems.
DiamondLightSource has released 'claude-sandbox', an open-source sandbox tool that safely supports the autonomous execution of coding agents such as Claude Code.
Anthropic has launched redesigned Claude Code projects in beta, replacing folder-based management with a conversation-driven interface where Claude coordinates parallel work threads.
Claude Code has launched a beta version of Projects, redesigned to be conversation-based rather than folder-based.
The HarnessTax study reveals that even with the same AI model, costs can differ by up to 2x depending on the harness choice, with the simple open-source harness Pi demonstrating competitive performance.
It has been reported that Claude Code fails to reuse the prompt cache for follow-up requests after a sub-agent terminates, resulting in the rewriting of nearly the entire context and incurring charges.
Anthropic has released early access to 'Claude Mods', enabling the execution of TypeScript-based custom features within Claude Code.
Nine Three.js game development skills usable in Codex and Claude Code have been released, providing self-verification capabilities based on a visual quality scorecard.
The 'Diagram Design' skill for AI agents supports 39 editable diagram types and applies minimal design principles based on HTML+SVG.
Anthropic engineers now ship 8x more code per quarter, with Claude authoring 80% of it, driving a 25x increase in CI jobs that required scaling test impact analysis.
Analysis of 17,000 agent runs reveals that Claude Code rarely performs web searches and that tool selection discrepancies among coding agents are frequent.
RTK's token savings metrics diverge from actual cost measurements, with the latest models showing negligible or even increased costs.
A researcher reverse-engineered the Firecracker MicroVM inside Claude Code Web, confirming the existence of Anthropic's private PaaS 'Antspace' and web app builder 'Baku'.
UiPath measured Claude Code skill activation rates and improved them from a baseline of 46.3% to 67.3% using unique marker-based descriptions.
Portfolio managers and developers are using Claude, Claude Cowork, and Claude Code for fundamental research and tool building.