Web Tools, Subagents, Contextual Hooks, and Per-File Code Review
0.8This release introduces web tools for searching and fetching content from the internet, enhanced hooks with new action types, subagents for parallel task execution, and improved supervised mode with per-file review capabilities.
Video transcript
The video shows the Kiro IDE with the AI chat panel open. The user asks "what is the latest version of node.js?", and the agent performs a web search (shown as a "Web search" card with "10 results") and answers that the latest version is v24.11.0 (LTS), with a bulleted list of Node.js 24 highlights and a cited source link to nodejs.org. The user then asks "get me the changelog for nodejs 24," and the agent uses a web-fetch tool — shown as a "Fetched" card with the nodejs.org release URL and a size/time indicator (24.5 KB · 478ms) — while a "Working" status appears at the bottom.
Kiro can now search the web and fetch content from URLs directly in chat. Use web tools to look up current documentation, find the latest library versions, or research solutions to technical problems. This keeps your development workflow in one place without switching to a browser.
Video transcript
The video shows a Kiro IDE hook configuration screen for a hook named "TS File Formatter," described as automatically running Prettier to reformat TypeScript files on save. The form displays an "Event" set to "File Saved," a watched file path of **/*.ts, an "Ask Kiro" action, agent instructions to run npx prettier --write, a "Hook enabled" toggle, and a "Delete Hook" button. The view then shows the hook in action: an AI chat session titled "Execute hook: TS File Formatter" and a related "optimize imports in gameplay.ts" session, with the agent responding to the triggered task.
Introducing contextual hooks with two new triggers: Prompt Submit and Agent Stop. These hooks fire at key moments in the agent workflow, letting you inject context or run commands before the agent acts. Choose between Agent Prompt actions to instruct the agent with natural language, or Shell Command actions to run commands locally without consuming credits.
Video transcript
The video shows the Kiro IDE with a code editor on the left and the AI chat panel on the right. The user asks the agent to "retrieve the latest 5 issues" from a GitHub repository (the Model Context Protocol registry) and use subagents to summarize each. The agent calls an MCP tool (list_issues), then shows several "Invoked Subagent" / "Invoking Subagent" cards, each stating it is "Delegating summarization of issue #[number] to a sub-agent for parallel processing." It then outputs a numbered summary of the five issues, each with an issue number, date, and description, while a "Working" status shows at the bottom.
Introducing subagents for parallel task execution. Kiro can now run multiple tasks simultaneously or delegate to specialized subagents. Two built-in subagents are available: a context gatherer for exploring projects and a general-purpose agent for parallelizing tasks. Each subagent has its own context window, keeping the main agent context clean. Use subagents to investigate multiple data sources in parallel, analyze GitHub issues across repositories, or extend your context window limits without requiring summarization.
Video transcript
The video shows the Kiro IDE with a tasks.md implementation plan open and the AI agent working on a documentation task. The agent reads several files and then pauses for review: the chat panel shows pending change cards labeled "Need input for edits to README.md" (+229 −56) and "Need input for creation of MIGRATION.md" (+321), with a purple "Waiting on your input" bar offering "Reject all" and "Accept all" buttons, plus per-change accept and reject controls and a "View diff" option. After the user accepts, the README.md card updates to "User accepted edits to README.md," and the editor reflects the applied changes. The Autopilot toggle is shown in the off position.
Supervised mode now offers granular control over code changes with per-file review capabilities. When Kiro makes changes to multiple files, you can review each file individually and selectively accept or reject changes. This turn-based approach works in both vibe chat and spec chat sessions, giving you full visibility into each modification.