Technology

Claude Code's Latest Updates

Claude Code rolls out Opus 4.8 as default, introduces dynamic workflows, security plugins, and performance enhancements for developers.

By Daniel Singer, Editor, StartupHub.ai

3 min read

Screenshot of Claude Code interface highlighting new features and model information.
A look at the recent enhancements and new default model in Claude Code.Credit: Claude Code Updates
Visual TL;DR
Claude Code UpdatesCore
Anthropic's developer tool gets significant new features
From the article 6 mentionsAnthropic's developer tool, Claude Code, continues its rapid evolution with a series of updates aimed at boosting developer productivity and AI model capabilities.
Opus 4.8 DefaultEffect
Most advanced model now default for various accounts
From the article 4 mentionsThe Week 22 updates, detailed on Claude Code Updates, see Claude Opus 4.8 becoming the new default for Max, Team Premium, Enterprise pay-as-you-go, and Anthropic API accounts.
Dynamic WorkflowsEffect
Orchestrate multiple sub-agents from a single script
From the articleThis week's rollout also introduces dynamic workflows, enabling the orchestration of numerous sub-agents from a single script.
Security PluginEffect
From the article 3 mentionsA new security-guidance plugin is designed to proactively review Claude's code modifications for vulnerabilities in real-time.
Performance GainsEffect
From the articlePerformance gains are evident with a fast mode now running on Opus 4.8, offering competitive pricing at $10/$50 per million tokens.
Boosted ProductivityOutcome
Enhancements aim to increase developer efficiency
From the articleAnthropic's developer tool, Claude Code, continues its rapid evolution with a series of updates aimed at boosting developer productivity and AI model capabilities.

Anthropic's developer tool, Claude Code, continues its rapid evolution with a series of updates aimed at boosting developer productivity and AI model capabilities. The latest digests highlight significant shifts, including the widespread adoption of its most advanced model.

The Week 22 updates, detailed on Claude Code Updates, see Claude Opus 4.8 becoming the new default for Max, Team Premium, Enterprise pay-as-you-go, and Anthropic API accounts. This model is configured for high effort by default, with an option for even higher intensity for complex tasks.

Key Features and Enhancements

This week's rollout also introduces dynamic workflows, enabling the orchestration of numerous sub-agents from a single script. A new security-guidance plugin is designed to proactively review Claude's code modifications for vulnerabilities in real-time.

Performance gains are evident with a fast mode now running on Opus 4.8, offering competitive pricing at $10/$50 per million tokens.

Previous weeks have seen further refinements. Week 21 brought Auto mode to the Pro plan, enhancing safety with background checks and expanding support to Sonnet 4.6. The `/usage` command now provides granular breakdowns of plan limit consumption by skill, sub-agent, and plugin.

The Agent view, introduced in Week 20, offers a unified screen for each Claude Code session, detailing running processes, pending approvals, and completed tasks. Additionally, the `/goal` command allows Claude to maintain context across turns until a defined completion condition is met.

Plugin management has been streamlined, with support for loading from .zip archives and URLs via `--plugin-dir` and `--plugin-url` respectively. Auto mode also gained hard deny rules for unconditional blocking of specific actions.

Windows users benefit from the removal of Git Bash as a requirement, with Claude Code now defaulting to PowerShell. The introduction of `claude ultrareview` brings cloud-based code review capabilities to CI pipelines and scripts.

A public research preview of `/ultrareview` offers a fleet of cloud-based bug-hunting agents that report findings directly to the CLI or Desktop. Custom themes and a redesigned web interface with a new sessions sidebar also enhance the user experience.

The introduction of Claude Opus 4.7 as the default for Max and Team Premium plans in Week 16, alongside an interactive `/effort` slider, allows for fine-tuning coding intensity.

Early previews of the Ultraplan allow for cloud-drafting and remote execution of plans. The Monitor tool streams background events, enabling Claude to react to logs in real-time.

Computer interaction via the CLI, a research preview from Week 14, enables Claude to launch native apps and navigate GUIs directly from the terminal.

Auto mode, in research preview since Week 13, automates permission prompts for safe actions while blocking risky ones.

Sources

TopicsClaude CodeAnthropicAI DevelopmentLLMDeveloper ToolsCode GenerationAI AgentsMachine Learning

Daniel Singer

Written by

Daniel Singer

Editor, StartupHub.ai

Daniel Singer is the editor of StartupHub.ai, a technology expert and thought leader on AI and its applications across sectors, from fintech and healthcare to developer tooling and consumer software. He writes and tests the tools covered here thoroughly and regularly, and built StartupHub.ai to give founders, operators and buyers a clearer read on what they are actually being sold.

More from Daniel SingerLinkedIn

Test your Agent Readiness now

Paste your website. How well ChatGPT, Claude and the other agents can read and use a site, in about a minute, free.

Watch

The videos behind the news, one flip at a time

The feed mixes the videos from our articles with startup demos and you flip through them like TV channels for free with no signup.

Open the video feed