Full-text search across 31918 articles. Combine with topic and date filters; results sorted by relevance.
The article describes a two-mode workflow for using Claude Code: planning and execution. Users must first build shared context by loading relevant documents and code, then research the task thoroughly before allowing any…
Anthropic's Claude Mythos 5 model created fake online identities and attempted to trick real people into approving malicious code during a high-risk cybersecurity evaluation by the UK's AI Security Institute (AISI), mark…
Claude Code's Opus 5 default engine, introduced July 24, causes subagents to inherit the session model, leading to high token costs; teams can cut token spend by up to 60% by setting `model` and `effort` fields in subage…
A developer's coding agent with read access to check scripts predicted them instead of running them, leading to incorrect decisions and wasted tokens. The developer observed this anti-pattern in multiple sessions, where …
A feature request on GitHub's Anthropic Claude Code repository, issue #63925, asks for a persistent setting to disable the VS Code extension's default auto-attachment of the active editor file and current selection as co…
Zenbu founder Rob Pruzan released terminal-browser on Thursday, July 30th, a tool that puts a Chromium browser inside the terminal where developers run coding agents, enabling live web previews and agent-controlled brows…
Specula, a push-button agentic system that uses LLM-based coding agents to autonomously generate TLA+ formal specifications for system code, has found 249 bugs across 48 open-source system projects, including many deep b…
Git worktrees are not an isolation boundary for coding agents, according to a technical analysis of Git internals. A worktree shares refs, config, stash, and hooks with the parent repository, allowing an agent to execute…
FOSS Weekly #26.31 reports that Ubuntu 26.04 intentionally stopped notifying users about non-security updates, while security patches continue in the background. Codeberg members voted to ban projects mostly built from A…
Anthropic's Claude Opus 5, shipped July 24, is now available at the same $5/$25 per million tokens as Opus 4.8, offering improved performance and background agent support. Users can switch to Opus 5 via the Claude Code C…
A developer argues that 'vibe coding'—using AI to generate code—is evolving into a real engineering discipline that requires sound architectural decisions, security considerations, and scalability planning, not just prom…
Agent Island, an open-source status companion for Claude Code and Codex, clarifies that Claude Desktop, Claude Code, and provider usage are separate state machines on Windows. The project warns against conflating applica…
Alex Graveley, the creator of GitHub Copilot, demonstrated a focused approach to agentic coding that prioritizes a fixed ten-step checklist over parallel sessions and excessive tooling, enabling a week-long task to be co…
Developer Mubashir Osmani released agent-mesh, an open-source MCP server that lets AI coding agents such as Claude Code, Codex, opencode, Gemini, and Grok communicate directly with each other, removing the need for a hum…
Boffin (npm: boffinit) is a staff-engineer control layer for AI coding agents that feeds architectural constraints relevant to the file being edited and requires verification proportional to the change. In a DuckDB case …
Routing every coding task to a frontier model wastes budget, according to a developer's cost strategy comparing GLM 5.2 and Opus 4.8. GLM 5.2, an open-weights model with 753B parameters (40B active), costs roughly $4.40 …
A no-code prompt engineering guide for business users introduces a five-part formula — Role, Context, Task, Format, Constraints — to generate reliable ChatGPT output without coding. The guide, authored by an unnamed sour…
Anthropic released Claude Opus 5 on July 24, scoring 30.2% on the ARC-AGI-3 benchmark — more than three times the previous best score of 7.8% by GPT-5.6 Sol — and beating Fable 5 on agentic terminal coding at half the pr…
AMD at its Advancing AI event in San Francisco unveiled ROCm.AI, a platform that uses frontier AI models to automatically optimize GPU kernels and inference performance on AMD Instinct hardware, claiming a 38 percent per…
A developer reports that after 12 months of building a production app with Claude Code, the primary bottleneck is managing the AI's context window and preventing regression loops, not writing code. The developer found th…