ClaudeDevs

Anthropic halved Sonnet 5.5 cache-read pricing on the Claude Platform to $0.10 per million tokens, with input at $2 and output at $10. API only, with no change to Claude Code usage limits.

Claude Code will detect an imminent 5-hour session limit mid-task and spend a small fixed allowance from your weekly limit to reach a clean stopping point instead of cutting off mid-edit, for Pro, Max, and Team Premium plans.

Claude Platform now exports OpenTelemetry traces from the Messages API, Managed Agents sessions, and the Claude SDKs. Spans follow gen_ai conventions with token counts, model, and tool names, and optional full prompt and tool content.

Anthropic added `ant apply` to the ant CLI. Teams can now declare Claude Managed Agent environments, agents, skills, memory stores, and deployments as files in a repository and keep the API's resources in sync with them. Docs are published on the Claude platform.

Anthropic engineers walk through a scheduled Managed Agent that reads Slack and GitHub and posts a daily brief. Covers vault credentials, memory stores, cron deployment, prompt-injection limits and budget caps, with failure modes and a reference repo.

Anthropic says Claude Code now gzips the conversation it sends on every turn so it reaches the model faster, and cloud sessions begin starting up as soon as you start typing. The post gives no measured latency numbers.

Claude Platform's Opus 5.5 fast mode exits research preview with no waitlist. It runs up to 2.5x faster at $8/$40 per million tokens, is available on Microsoft Foundry, with Google Cloud coming. Claude Code fast mode is unchanged.

The Claude Code desktop app gains multi-session features: reusable environments with optional worktrees, nested sessions for parallel tasks reporting back to the main thread, suggested follow-ups, side-question tabs, and a sidebar showing which sessions are working or waiting.

Anthropic's playbook prices Claude Code tasks on Opus 5.5 using list prices and illustrative sessions. It explains turn count, cache behavior, effort levels, subagent model choice, and compaction, and includes a calculator and steps to measure your own sessions with /usage.

Claude Code Projects, a beta feature where Claude coordinates ongoing work by running each task as its own parallel thread, is now open to all waitlisted Pro and Max users, with a walkthrough and docs linked.

The Claude Code desktop app now lets you drag any pane, like the diff or terminal, into its own window across screens while Claude keeps working, and you can run multiple sessions side by side or stacked.

Anthropic's guide to Claude Sonnet 5.5: when to choose it over Opus 5.5, pricing and model details, a six-step migration from Sonnet 5 covering breaking API changes, and tuning advice on effort levels, caching and progress updates. The X thread collection was incomplete and the video was not read.

Anthropic moved Code Review out of research preview for Team and Enterprise. Reviews run about 3x faster, findings state the problem, trigger, and fix, and admins pick Quick, Standard, or Deep per repository. Enterprise orgs can claim promotional credits through January 15, 2027.

Anthropic's developer team shares a working playbook for Claude Opus 5.5: delegate full tasks with clear done-criteria, skip 'think carefully' since it always reasons first, and check in after long autonomous runs.

Anthropic ships Sonnet 5.5, 30% faster than Sonnet 5 at unchanged pricing, with always-on thinking, effort guidance, and org-bound preserved thinking that resets when accounts switch mid-session.

Anthropic open-sourced Claude Commerce Agents, a blueprint for shopping and merchant agents with reference implementations across retail, travel, telecom and entertainment, and a Claude Code plugin that scaffolds an agent against an existing backend. The post covers architecture, latency and cost te

Anthropic's Python and TypeScript SDKs now include computer use and browser use toolsets. The SDK runs the action loop and routes Claude's requested clicks and keystrokes to drivers instead of custom code.

Anthropic describes how its team uses Claude Tag for on-call: when a Slack alert fires, Claude pulls metrics, diffs recent deploys, checks feature flags, identifies a likely cause, and proposes a fix for a human to approve and merge.