17 November 2026 · 3 min read
AWS Launches Kiro Bots: AI Agents That Take On Specific Development Roles
AWS has officially launched Kiro bots, long-running AI agents that handle tasks like code review, bug triage, and release preparation, letting developers focus on higher-level decisions.
What was announced
On 17 November 2026, AWS officially launched Kiro bots within Kiro Crew, its AI agent-based development workspace [1][2].
What Kiro bots bring
Kiro bots are long-running AI agents that take charge of specific parts of the software development process. Unlike one-off AI prompts, each bot maintains its own name, chat window, workspace, memory, and permissions [1][2]. They can run multiple tasks in parallel, track progress, and adjust direction when needed. If a task requires a developer’s decision, the bot asks only about that item while continuing other unaffected work [1][2].
This design addresses a common pain point: when several AI agents are assigned tasks simultaneously, developers previously had to remember each session’s progress, catch up on results, and manually find pending decisions. Kiro bots solve that by giving each agent ongoing task-management capabilities and isolated context [1][2].
Four built-in roles
AWS outlined four roles that Kiro bots can perform out of the box [1]:
- Issue intake: Reviews and categorises new bug reports and feature requests, flags duplicates, and writes reproduction steps.
- Build status checks: Periodically checks CI environments, distinguishes new errors from known transient ones, and alerts developers only to issues that need attention.
- Code review: Reviews code changes against team coding rules and security standards, and suggests revisions.
- Release preparation: Verifies that code changes have been tested and release notes written, then asks for final approval before release.
The service also supports scheduling repetitive tasks to run automatically [1].
Developer control and model choice
Developers can configure each bot individually. They can edit what a bot remembers, adjust its rules, and grant only the tools and permissions needed for its duties [1][2]. For example, a research bot can read files and search the web but cannot modify files; a deployment bot can run scripts but requires developer approval for changes [2].
Memory is isolated per bot, so what one bot learns does not affect another. Each bot improves over time through developer feedback and task outcomes. Developers can view, edit, or delete everything a bot remembers [2].
Developers also choose the AI model for each bot. A higher-performance model can be used for complex planning, while a faster, lower-cost model handles routine checks, balancing cost and performance [1][2].
Availability and ecosystem
Kiro bots run wherever Kiro Crew is installed: on a developer’s laptop, a Mac mini, or an Amazon EC2 instance [1][2]. Kiro Crew itself is an open-source workspace released under the Apache License 2.0 and supports macOS, Linux, and Windows [1]. Kiro was first unveiled in July 2025 [2].
No pricing details have been announced for the bots themselves; they operate within the Kiro Crew environment, which is free and open source.
What it means for developers
Kiro bots shift the developer’s role from managing individual AI tasks to overseeing a team of specialised agents. Instead of repeatedly prompting an AI for each small job, developers can assign ongoing responsibilities to bots and step in only when human judgement is required. This lets teams focus on architectural decisions, final reviews, and strategic work while the bots handle repetitive checks, triage, and readiness tasks [1][2].
For teams already using AI coding assistants, Kiro bots offer a way to scale AI usage without multiplying the overhead of context switching and manual tracking. The isolated memory and permissions mean each bot can specialise and improve without interfering with others, making it practical to delegate distinct parts of the development lifecycle to AI.
Sources
- "Find Bugs and Review Code"... AWS Launches ‘Kiro Bots’ for Developers - First-Class Business Newspaper Financial News
- AWS launches Kiro bots to manage roles across software development
- aws
- kiro bots
- ai agents
- developer productivity
- code review
Researched and drafted with AI from the sources above.
Learn to build with AI
Hands-on tutorials on Cursor, Claude Code, AI agents and more.