LIVE
⚡Model Context Protocol Architecture: Tools, Resources & Claude•⚡OpenCode Agents: Architecture, Subagent Delegation & Rules•⚡OpenCode Agents Guide: Multi-Agent Patterns and Structure•⚡OpenCode Intro Docs: Provider Hub, LSP Engine & Tool Stack•⚡OpenCode Console & TUI Setup: Intro Docs Walkthrough•⚡Configuring Custom Instructions for OpenCode with AGENTS.md•⚡Model Context Protocol Architecture: Tools, Resources & Claude•⚡OpenCode Agents: Architecture, Subagent Delegation & Rules•⚡OpenCode Agents Guide: Multi-Agent Patterns and Structure•⚡OpenCode Intro Docs: Provider Hub, LSP Engine & Tool Stack•⚡OpenCode Console & TUI Setup: Intro Docs Walkthrough•⚡Configuring Custom Instructions for OpenCode with AGENTS.md•⚡Model Context Protocol Architecture: Tools, Resources & Claude•⚡OpenCode Agents: Architecture, Subagent Delegation & Rules•⚡OpenCode Agents Guide: Multi-Agent Patterns and Structure•⚡OpenCode Intro Docs: Provider Hub, LSP Engine & Tool Stack•⚡OpenCode Console & TUI Setup: Intro Docs Walkthrough•⚡Configuring Custom Instructions for OpenCode with AGENTS.md•⚡Model Context Protocol Architecture: Tools, Resources & Claude•⚡OpenCode Agents: Architecture, Subagent Delegation & Rules•⚡OpenCode Agents Guide: Multi-Agent Patterns and Structure•⚡OpenCode Intro Docs: Provider Hub, LSP Engine & Tool Stack•⚡OpenCode Console & TUI Setup: Intro Docs Walkthrough•⚡Configuring Custom Instructions for OpenCode with AGENTS.md•
ActiveUnified agentic development environmentSource check · September 13, 2026

Kiro

A shared agent harness across IDE, CLI, web and mobile; Amazon Q Developer CLI has a documented migration path to Kiro CLI.

Code EditingRepo UnderstandingTerminalMcpSubagentsBackground RunsPull RequestsSkills And Plugins

Product briefing

What it is and how it fits a coding workflow

Kiro is an agentic development environment organized around one harness that is available through several clients. The IDE adds editor integration and specification-driven planning; the CLI supports terminal and headless work; Kiro Web runs repository tasks in a managed sandbox and can prepare pull requests; mobile is for following sessions and reviewing work. AWS documents Kiro CLI as the next update of Amazon Q Developer CLI and provides a migration path. Existing users can continue using the old `q` commands, so treat Kiro as the recommended successor path rather than an unrelated second CLI agent or an instantaneous removal of Q CLI. The `.kiro` project configuration carries guidance such as steering rules, custom agents, hooks, MCP servers and skills across surfaces, while each client has different permissions and execution location. That makes Kiro a useful option for teams that want structured requirements and a continuous handoff from local planning to cloud implementation. It is not an open-source, provider-neutral runtime: model selection follows the product’s current catalog and account or enterprise geography policies. Pay attention to where a session runs—local IDE/CLI work can touch the developer machine, whereas web sessions operate in a cloud sandbox connected to selected repositories. The cloud interface can open a PR, but documentation says changes are reviewed before reaching the default branch. Before enabling it for a company repository, validate identity setup, data residency, repository grants, audit requirements and how secrets are handled. Compare Kiro with an IDE-only agent when local control is more important than shared cross-surface sessions.

Best fit

  • Teams wanting specifications, rules and reusable agent setup across several clients
  • Moving a task between local IDE/CLI sessions and cloud pull-request work
  • Engineering groups that need hooks, MCP, subagents and centrally managed accounts

Probably not your fit if…

Offline-only coding, a fully self-hosted open-source harness or unrestricted choice of arbitrary inference endpoints.

What stands out

  • ↗ One documented harness spans IDE, CLI, web and mobile
  • ↗ Specs, steering, hooks, MCP, skills and subagents are part of the workflow
  • ↗ Cloud sessions provide sandboxed repository tasks and reviewable PR delivery

Trade-offs to check

  • ! Model catalog, plan and region affect availability and data path
  • ! Local and cloud surfaces have different permission and execution boundaries

Primary references

Verify features at the source

This profile is an editorial comparison, not a performance benchmark. Product features, plans and availability can change; verify your required workflow before adopting it.

Related field notes

Further reading

Explore related tools

Other profiles for similar workflows

Search all coding agents →