AI Briefing#coding-agents
The latest AI and developer news about #coding-agents, with the original source and a short summary.
The latest AI and developer news about #coding-agents, with the original source and a short summary.
UC Berkeley researchers identified the 'Harness Tax' phenomenon, where achieving the same quality with coding agent harnesses can cost up to 5 times more depending on the harness design.
Released a coordinator-worker based 'Chief of Staff' pattern and operational loop to address context loss and lack of verification in Claude Code.
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.
Cloudflare has open-sourced a security audit skill for AI coding agents to support automated vulnerability detection.
The AI SDK harness layer now supports native subscription authentication without requiring code changes.
Analysis of 17,000 agent runs reveals that Claude Code rarely performs web searches and that tool selection discrepancies among coding agents are frequent.
The AI SDK harness layer now supports GitHub Copilot via an official adapter, enabling coding agent switching without code changes.
The open-source tool 'Portable Resume' has been released to transfer session context between various coding agents such as Claude Code and Codex.
Inception Labs has released Mercury 2.5, which increases intelligence by 40% and generates 1,107 tokens per second.
OpenAI revealed its internal research automation status, stating that agent labor has surpassed human labor.
PyTorchKR selected this week's papers, introducing recent research on enhancing agent capabilities and interpreting internal structures, such as long-horizon autonomous coding agents, recursive self-improvement, and the approximation of symbolic structures in neural networks.
OpenAI reported that in August 2026, agent labor within its research organization surpassed human labor, leading to strengthened safety controls and a temporary pause in RL training as RSI progressed.
A study on coding agents analyzes the preference for grep over semantic navigation (LSP) and identifies the impact of tool output formats on agent performance.
Analysis of 17,000 runs revealed that Claude Code primarily relies on prior knowledge while Codex mainly uses web search, showing distinct tool selection tendencies.
Hugging Face has released 'funes', a local open-source memory tool that preserves cross-session reasoning for coding agents like Claude Code.
A story of regaining LLM/Python instincts and codebase understanding through hand-coding
Released a pre-critique skill to prevent agents from over-complying.
A local agent memory engine that re-weights memories based on failure/success feedback.
As coding agents take on deployment and operations, infrastructure itself is being reorganized around agents.
Continue has expanded beyond an IDE extension into an agent platform for PR verification.