Anthropic's new rules for "context design in the Claude 5 era," published July 24, 2026. Cutting 80% of Claude Code's system prompt caused no drop in accuracy. This piece lays out six shifts—lightening CLAUDE.md and moving procedures into skills—along with the concerns around automatic memory and de
MCP, the connectivity standard from Anthropic, reaches its biggest revision since launch on July 28, 2026. It removes "sessions" from the core to go stateless, and adds two extensions: "Tasks," a common convention for long-running work, and "MCP Apps," which return a UI. Breaking changes are include
Claude Opus 5, the flagship model Anthropic just released, is now available across 10 GitHub Copilot surfaces — VS Code, the CLI, cloud agents, and more. Pricing moves from multipliers to usage-based billing, organizations must explicitly enable it, and security-adjacent requests may be blocked.
Claude Code v2.1.219 restores nested subagents — turned off by default just two days earlier — to "depth 3 by default," and introduces a new "fewer than 15" guideline for dynamic workflows. A look at the tug-of-war between autonomy and safety, and where these changes matter most for unattended opera
The latest release of GitHub Copilot CLI can now read the vendor-neutral "Open Plugin Spec" and MCP configuration files. Extensions built for other agents can be carried over as-is, while the default settings for the OS sandbox have moved toward tightening.
OpenAI is rolling out voice control to the ChatGPT desktop app. With full-duplex GPT-Live, you can run multiple Codex/Work agents by voice alone. It speeds up the workflow, but brings new slip-ups too: the ambiguity of spoken instructions and drift in transcription.
Anthropic released Claude Opus 5 on July 24. At the unchanged $5/$25 pricing, it delivers performance approaching the publicly top-tier Fable 5 and replaces the default Opus in Claude Code. Here's a rundown of the shifts in pricing, benchmarks, and agent operations—along with the caveats to keep in
On July 22, Cursor announced Cursor Router, which automatically selects the AI model to use for each request. It's said to be up to 50% cheaper than pinning to Opus, but concerns remain — such as it being hard to see which model wrote your code. We break down the cost, mechanics, and caveats.
OpenAI's Codex 0.145.0 is out. Its /import can now pull in Cursor and Claude Code MCP servers and settings wholesale, lowering the barrier to switching. Multi-Agent V2 also reaches a stable release.
Claude Code v2.1.218 looks like a bundle of minor fixes, but a single thread runs through it. /code-review now runs in the background, /deep-research launches only when invoked manually, and fork skills default to running behind the scenes—pushing heavy automated work out of the conversation so noth
On July 21, Google refreshed three Gemini Flash models. It again held back the flagship 3.5 Pro, instead strengthening the mid-tier — including a 3.6 Flash that cuts output tokens by 17% — to target "the cost of keeping agents running." A developer-focused rundown covering price tags, benchmarks, an
Announced by AWS on July 20, CloudWatch Coding Agent Insights lines up the consumption, cost, and output of Claude Code, Codex, and Copilot on a single screen. This piece weighs both sides — the benefit of measuring adoption gains in hard numbers, and the danger of repurposing individuals' line coun
