Brian Beach

In case you missed it, @kirodotdev now supports Claude Opus 4.6 and Claude Sonnet 4.6 with 1M context window. I'm excited to try the this for modernization of large code bases. How are you going to use it? #Kiro

In case you missed it, @kirodotdev has an all new terminal user interface (TUI). You can try the new usr experience by starting Kiro with `kiro-cli --tui`. How could Kiro make the TUI better for you? #Kiro

In case you missed it, @Kirodotdev enterprise plans now include access to open weight models including Deepseek, Minimax, and and Qwen. Which open weight models would you like Kiro to add? #Kiro

In case you missed it, @kirdotdev launched Model Governance — enterprise admins can curate which AI models are available, set defaults, and enforce data residency requirements. Which models do you trust? #Kiro

In case you missed it, @kirodotdev launched MCP Registry Governance — enterprise admins can now control which MCP servers are approved across their org with a hosted JSON registry. Which MCP servers do you trust? #Kiro #MCP

In case you missed it, #Kiro CLI launched File References — type @path to inject file contents or directory trees directly into your message. @path saves you the round trip with no tool calls needed. What files would you pull in first?

In case you missed it, #Kiro launched Tool Context Insights — /tools now shows estimated token counts per tool so you can see what's eating your context window. I've been using it to spot MCP servers that are bloating my context. Do you track your token usage?

In case you missed it, #Kiro CLI launched Dynamic Model Selection — switch models on the fly with /model, tab completion, and fuzzy matching. I've been jumping between models depending on the task complexity. How often do you switch models mid-session?

In case you missed it, #Kiro launched MCP Prompts — MCP servers can now surface prompts, resource templates, and request input from you directly in chat. I'm using it to pull prompts from a prompt library. How are you going to use it?

In case you missed it, #Kiro launched Bugfix Specs — describe a bug and Kiro walks you through root cause analysis, fix design, and regression prevention before writing any code. Do you create a plan for your bug fixes?

In case you missed it, #Kiro launched Design-First Specs. Some features don't need user stories and I want start with the technical design. Now I can. Do you think every change needs a requirements documents?

In case you missed it, #Kiro updated the telemetry and metrics to standardize user activity reporting across the IDE, CLI and plugins. You can now track credit consumption and overages for individual users. Which metics do you want to see added?

In case you missed it, @Kiro IDE and CLI now support enterprise authentication using #Okta or #Entra ID in addition to #AWS IAM Identity Center. Which identity provider do you want added next?

In case you missed it, #Kiro launched Web Tools Governance — enterprise admins can now control web search and fetch access across their organization from the AWS console. How does your team handle AI tool governance?

In case you missed it, #Kiro launched Agent Skills. A simple, open format for giving agents new capabilities and expertise originally developed by Anthropic. I'm excited to leverage the growing catalog of skills already out there in the world. What would you turn into a skill?

@Kirodotdev is now available in AWS GovCloud (US) 🛡️ Same agentic, spec-driven dev experience, built for regulated workloads and elevated security standards. Learn more here: kiro.dev/blog/introduci…

In case you missed it, @Kiro added support for Claude Opus 4.6 — Anthropic's most powerful model and the world's best for coding. I have found that Opus 4.6 is up to date on the latest technologies. What are you going to build with Opus 4.6?

In case you missed it, you can now choose from three open weight models in the @Kiro IDE and CLI. Individual accounts now include access to #DeepSeek, #MiniMax, and #Qwen models that are fast and cost-effective. What is your favorite open weight model?

In case you missed it, #Kiro launched Custom Subagents — define your own specialized agents with custom system prompts, model selection, and tool access. What kind of agent would you build?

In case you missed it, #Kiro CLI launched subagent access control. This is great for teams that want to lock down which custom agents can run in production environments. How are you managing agent permissions?