Kiro-KitPreset Reference
Every preset that ships with kiro-kit, with its full command, agent, and skill catalog.
Every table in this section is generated from presets/*/manifest.json and the
files those manifests declare. If it is listed here, it ships.
Presets
| Preset | Commands | Agents | Skills | Workflows | Category |
|---|---|---|---|---|---|
ai-engineer | 70 | 20 | 37 | 4 | ai-engineer |
backend | 67 | 25 | 23 | 4 | backend |
data-ai | 70 | 20 | 31 | 4 | data-ai |
devops | 66 | 26 | 25 | 4 | devops |
frontend | 71 | 20 | 21 | 4 | frontend |
fullstack | 74 | 26 | 29 | 4 | fullstack |
kiro-kit-dev | 66 | 19 | 22 | 4 | kiro-kit-dev |
mobile | 71 | 23 | 26 | 4 | mobile |
sa | 69 | 26 | 31 | 4 | sa |
Totals
| Artifact | Total across all presets |
|---|---|
| Presets | 9 |
| Commands | 624 |
| Agents | 205 |
| Skills | 245 |
Presets are self-contained, so these totals count each preset's own copy of an
artifact separately. Two presets shipping a code-reviewer agent count as two.
Kiro Powers by preset
Each preset recommends curated Kiro Powers grouped by priority tier.
| Preset | Essential | Recommended | Optional |
|---|---|---|---|
ai-engineer | AWS Infrastructure as Code, Terraform | Context7, Exa, Hugging Face | Langfuse, LangSmith, Datadog, Pinecone, Snyk |
backend | Supabase | Neon, Postman, Context7, Upstash | Stripe, Snyk, Sentry |
data-ai | ClickHouse | Context7, Exa, Hugging Face | Neon, New Relic, Weights & Biases, Pinecone, LangSmith |
devops | Terraform | Datadog, Snyk, Depot, Context7 | Harness, AWS CDK, Pulumi, Grafana |
frontend | Figma | Netlify, Vercel, Context7, Storybook | Sentry, PostHog, ScoutQA |
fullstack | Supabase | Figma, Netlify, Stripe, Context7, Clerk | Firebase, LaunchDarkly, Sentry, Resend |
kiro-kit-dev | Context7 | Snyk | Postman |
mobile | Firebase | Figma, Context7, Expo | ElevenLabs, Bria, RevenueCat, OneSignal, Sentry |
sa | Terraform, AWS CDK | Context7, Figma | Datadog, Snyk |
Example specs by preset
| Preset | Worked example specs |
|---|---|
ai-engineer | AgentCore Support Chatbot · MCP Server On Lambda · Multi Agent Orchestrator · RAG Knowledge Assistant |
backend | API Key Authentication · Background Job Queue · Caching Strategy · Webhook Delivery System |
data-ai | Customer Churn Prediction · LLM Evaluation Harness · Model Evaluation Pipeline · RAG Chatbot |
devops | AWS CDK Infrastructure · AWS Serverless API · Blue Green Deployment · CI CD Pipeline |
frontend | Analytics Dashboard · Landing Page · Product Listing Page · Web Performance Optimization |
fullstack | Stripe Checkout |
kiro-kit-dev | Add CLI Command · Add New Preset · Spec Library Expansion |
mobile | Offline First Notes |
sa | Data Lake Architecture · Event Driven Microservices · Multi Region DR · Three Tier Web Architecture |
Minimum thresholds
Every preset must meet these minimums. They are enforced by
preset-thresholds.test.ts,
and this table is read from that test file.
| Artifact | Minimum |
|---|---|
| agents | 16 |
| skills | 22 |
| commands | 40 |
| hooks | 6 |
| workflows | 4 |
Native Kiro Agent Hooks
Event-driven automation that runs inside Kiro, shipped disabled by default in every preset.
ai-engineer
AI Engineer kit for building production chatbots and agents on AWS with 20 agents, 37 skills, 70 commands — Amazon Bedrock AgentCore (Runtime, Gateway, Memory), MCP server hosting, Strands Agents, Bedrock Knowledge Bases RAG, guardrails, and agent evaluation