Claude Code new feature /ultraplan: Create an implementation plan on the web, then run it with one click after editing

ChainNewsAbmedia

Anthropic product lead trq212 announced on April 10 that Claude Code has added the /ultraplan feature, letting developers create a complete implementation plan in the web interface, edit and adjust it, and then choose to run it on the web or return to the terminal. The preview is already available to all users who have enabled Claude Code on the web.

Plan in the cloud, run locally

The core idea behind /ultraplan is to separate “planning” from “execution.” trq212 explains that implementation sometimes requires a local environment and interactivity, but planning can be done in the cloud because it mainly involves reading code and understanding intent.

This addresses a common pain point for Claude Code users: when using /plan in the past, the plan was generated in the terminal, making it difficult to get an overview and edit. /ultraplan moves the plan to the web interface, providing a better visualization and editing experience, so developers can fully review and adjust the plan before executing.

Differences from /plan

The number of tokens used by /ultraplan and its subscription rate limits are roughly the same as /plan, and it won’t consume additional usage. The main difference is the user experience: the web interface offers better capabilities for reading, editing, and sharing.

The community’s response has been enthusiastic, with some developers saying this makes Claude Code “no longer feel like a black box, but more like a senior developer companion.” Others have complained that all usage is used up within the first 30 minutes—showing that demand is very strong.

This article Claude Code new feature /ultraplan: Build an implementation plan on the web, edit it, and run it with one click first appeared in Chain News ABMedia.

Disclaimer: The information on this page may come from third parties and does not represent the views or opinions of Gate. The content displayed on this page is for reference only and does not constitute any financial, investment, or legal advice. Gate does not guarantee the accuracy or completeness of the information and shall not be liable for any losses arising from the use of this information. Virtual asset investments carry high risks and are subject to significant price volatility. You may lose all of your invested principal. Please fully understand the relevant risks and make prudent decisions based on your own financial situation and risk tolerance. For details, please refer to Disclaimer.

Related Articles

Gate founder Dr. Han: Reshaping the platform’s capability boundaries with an AI ecosystem and multi-asset strategy

Gate founder Dr. Han introduced the platform’s development in the anniversary public letter, including deepening the AI ecosystem and a multi-asset trading framework. Gate is building an AI product ecosystem and has launched a TradFi trading section, covering multiple asset types to meet users’ needs. In addition, it has rolled out a digital Pre-IPOs participation mechanism, giving users a new way to access high-quality public offering projects. Going forward, Gate will continue to strengthen its infrastructure and enhance users’ global digital asset trading experience.

GateNews3h ago

A Karpathy-inspired CLAUDE.md Breaks 15K Stars: How a Markdown File Can Tame an AI’s Bad Habits of Writing Code

A Markdown file named CLAUDE.md, based on observations and principles provided by Andrej Karpathy, helped the Claude Code project break 15,000 stars on GitHub. This document sets behavioral guidelines for common mistakes LLMs make when writing code, emphasizing goal-driven behavior, confirming requirements, and proactively disclosing trade-offs. It reflects developers’ focus on engineering AI behavior, and may even be more effective than choosing a more powerful model.

ChainNewsAbmedia7h ago

Starbucks CEO Reveals an “AI Barista Assistant”: Real-Time Help Brewing and Fixing Confusion in Channel Orders

Starbucks CEO Brian Niccol unveiled two AI systems at the Dreamforce conference: “Green Dot” and “Smart Q.” The former acts as a real-time assistant for store employees, while the latter solves order mix-up issues to improve service efficiency. In the future, the Starbucks app will be able to predict what customers order, enabling more convenient drink preparation. This series of AI deployments is intended to strengthen Starbucks’ role as a social space for people.

動區BlockTempo18h ago

Bitcoin miners face tough 2028 halving profits, as mining companies deploy AI infrastructure early to address the energy crisis

Bitcoin mining companies are facing profit pressure after the halving and challenges in power supply. Major operators are shoring up capital management through asset sales and seeking to diversify their energy-security arrangements. In addition, miners are turning to artificial intelligence infrastructure to improve operational flexibility. In the future, competitive advantage will belong to companies that can effectively manage debt and ensure long-term power supply.

ChainNewsAbmedia18h ago

V God shares: How I built a fully local, private, and self-controlled AI working environment

Vitalik Buterin proposed a local AI architecture, emphasizing privacy, security, and self-sovereignty, and warned about the potential risks of AI agents. He suggested avoiding cloud-based models and setting five major security goals to protect personal data. Testing showed that the NVIDIA 5090 laptop is the best hardware choice, and he stressed that a local-first strategy is crucial to the security of today’s AI tools.

CryptoCity19h ago

V God shares: How I build a fully local, private, and self-controlled AI working environment

Vitalik Buterin proposes a local AI architecture, emphasizing privacy, security, and self-sovereignty, and warns about the potential risks of AI agents. He recommends avoiding cloud-based models and setting five major security goals to protect personal data. Testing shows that the NVIDIA 5090 laptop is the best hardware choice, and he stresses that a local-first strategy is crucial for the safety of today’s AI tools.

CryptoCity23h ago
Comment
0/400
No comments