正文 · AI 翻译
Thanks to IT Home reader 惊讶仙人 for the tip!
IT Home reported on November 7 that Amp, an AI coding collaboration tool, announced Spaces yesterday (November 6), aiming to create a new way of working for AI agents and team collaboration.
Users no longer need to manually create, name, and manage multiple Agent threads. They only need to talk to one primary Agent, which understands their needs, breaks down tasks, and automatically creates and coordinates other Agent threads. IT Home has included the relevant screenshot below:
Amp noted that in traditional multi-Agent workflows, users often have to decide for themselves how to split up, assign, and track tasks. Spaces hands this organizational work to the Agent: each Space is a conversation between a user and a single Agent, which starts and coordinates multiple parallel threads for the user.
For example, a team can ask in a Space to “move a feature toward launch” or “conduct a security review.” The primary Agent turns the goal into several specific tasks, creates a separate thread for each, and displays progress as real-time cards in the chat.
Spaces is more than a tool for individual automation. It is designed as a place where teams can plan and direct work together: members can raise requests, report problems, or add ideas in the same chat; each task that is split off shows its current status and whether human intervention is needed.
Users can click a task card to enter the corresponding thread, review the Agent’s output, continue iterating, or move directly toward delivery. Team members can also discuss matters in comments under messages; Amp reads relevant comments only when someone mentions it with @amp, reducing unnecessary interruptions.
Amp positions Spaces as a “higher-level” approach to multi-Agent collaboration. Instead of directing Agents one instruction at a time, users describe their goals to an Agent capable of coordination, which then takes responsibility for breaking down tasks, creating threads, organizing progress, and summarizing results.
Official examples show that one member used a single Space to launch a service account feature; another scenario involves regular security reviews. To get started, users only need to click “+” in the Spaces section of the sidebar, set a title and project, and tell the Agent what work to handle.
This feature reflects a shift in AI coding tools from “individual question-and-answer” use toward team-based work and task orchestration. When Agents can independently break down and execute tasks in parallel, developers’ roles may shift more toward defining goals, reviewing results, and making key decisions, rather than manually scheduling every subtask.
Amp is an AI coding Agent and development environment for software development. Its core aim is to let developers tell the Agent “what to do,” while the Agent takes responsibility for reading and modifying code, running tests, iterating, and continuing to carry out tasks in the cloud or locally.
Amp can run tools and Shell commands in a user’s workspace to inspect code, run tests, and iterate quickly. It can also use the Task tool to assign specific subtasks to sub-Agents.