Google has expanded internal access to Anthropic's Claude Opus 5 for coding through its Antigravity development platform, while keeping Gemini as its primary and foundational model for internal develo…
A DEV Community post argues that buyers should stop treating "best AI writing tools" as a search for a generic blog generator, framing the category as one that requires more specific evaluation criter…
A developer using Claude Code on a Max subscription documented how making the top model (Fable) the session default caused built-in skills like /code-review and /simplify to inherit that model and exh…
Duolingo software engineer Sarah Deitke said 95% of the company's engineering organization reported learning something new and would recommend its lab-style AI training, which her DevEx AI team uses t…
AI coding agents such as Claude Code, GitHub Copilot, Cursor, and Codex can generate changes like 27 files and 1,800 lines of code in minutes, shifting software development's bottleneck from writing c…
A developer released Solo Stack, a set of 26 Claude Code skills for freelancers covering the client lifecycle from lead research and proposals to invoicing and weekly reviews, priced at $19 during lau…
AI-generated code that skips proper verification costs the software industry at least $1.17 trillion, according to a new report that tallies the expense across six categories including code generation…
Anthropic published a Claude Code study map generated from the Claude Code documentation at code.claude.com/docs, last updated September 21, 2026. The map organizes Claude Code concepts into stages nu…
A new MIT-licensed tool called never-again converts each bug a developer fixes into either a blocking hook or a one-line rule so AI coding agents stop repeating the same mistakes. The tool runs locall…
A developer's attempt to render AI agent state (working, blocked, idle) directly into the tmux window list failed on both candidate signals: Claude Code sets its pane title once and never updates it, …
A full stack developer compared Claude Code, Codex, and Cursor, concluding that no single AI coding tool is universally fastest and that each fits a different part of the development loop. The develop…
A developer analyzed 35 days of Claude Code MCP connection logs covering 27,257 connections and found that OAuth token refreshes inside the connection window make sessions 5.8× slower (3,170 ms median…
A developer built a local long-term memory layer for AI coding agents like Claude Code, Cursor, and Codex, moving memory out of the context window and onto disk so agents can query it via tools. The w…
Developer AEndrix03 released Graft, an open-source local-first persistent memory tool for AI coding agents that stores fixes, decisions, and project knowledge in SQLite with sqlite-vec and FTS5 and su…
An open-source project called Graft provides local-first persistent memory for coding agents, letting them store and recall fixes, decisions, and project-specific gotchas across sessions. The C11 CLI …
A developer published a minimal Python patcher that disables the Claude Code VS Code extension's automatic file-context attachment by nulling the argument to applySelectionUpdate in webview/index.js, …
A developer-run experiment found that 26 AI agents — spanning a 4-bit quantized 7B model up to Opus 5 — all passed the visible test suite for a seeded greedy-regex bug while all 26 remained broken, wi…
Thurbeen released Thurbox, an open-source tmux-based TUI and headless CLI for running multiple coding-agent CLIs side by side, each in its own persistent tmux session and git worktree. Thurbox launche…
Plasma built Radio, a shared communication channel that lets AI agents from different providers — including Claude Code, Codex, Cursor, OpenCode, and Grok — talk to each other and to humans in real ti…