← Back to all announcements
★★☆☆☆ 25/06/2026

Kiro achieves FedRAMP High and DoD IL-4/5 authorization in AWS GovCloud (US)

Federal agencies and DoD programs can now use Kiro's agentic AI coding platform for sensitive workloads—compliantly, inside GovCloud.

View original announcement →

Visual Summary

graph TD A{{Kiro in AWS GovCloud}}:::announced B((Federal Developers)):::external C(Amazon Bedrock):::compute D([Spec-Driven Development]):::feature E([FedRAMP High / DoD IL-4/5]):::feature F(Claude Sonnet 4.5):::compute G([MCP Integration]):::feature H([Data Isolation]):::feature I(AWS IAM / SSO):::compute B ==>|"authenticates"| I I ==>|"grants access"| A A -->|"generates specs"| D A -->|"invokes inference"| C C -->|"runs model"| F A -.->|"connects resources"| G A -->|"enforces"| E A -->|"ensures"| H classDef announced fill:#ff9900,stroke:#ec7211,color:#fff,font-weight:bold classDef compute fill:#e3f2fd,stroke:#1565c0,color:#1565c0 classDef storage fill:#e8f5e9,stroke:#2e7d32,color:#2e7d32 classDef feature fill:#fff3e0,stroke:#e65100,color:#e65100 classDef external fill:#f5f5f5,stroke:#616161,color:#616161

What's New

Kiro, AWS's agentic AI development platform combining an IDE and CLI, has achieved FedRAMP High and DoD Cloud Computing Security Requirements Guide (CC SRG) Impact Level 4 and 5 authorization within AWS GovCloud (US) Regions. This makes Kiro available to federal agencies, defense organizations, and public sector enterprises that handle sensitive and classified workloads requiring the highest levels of cloud security compliance. Organizations can now leverage Kiro's spec-driven, agentic engineering capabilities for mission-critical software development without compromising their regulatory obligations.

How It Works

  • Spec-driven development: Kiro transforms natural language prompts into structured requirements, architectural designs, and sequenced implementation tasks, then generates working code, documentation, and tests from those specs.
  • Agentic IDE and CLI: Developers interact through a standalone IDE or command-line interface; in GovCloud, VS Code plugin and inline suggestions are disabled, and the Autonomous Agent mode is not available.
  • Foundation model: Claude Sonnet 4.5 is the default (and currently only) foundation model in AWS GovCloud (US); automated model selection (Auto mode) is disabled at launch.
  • Cross-Region Inference (CRIS): Customers in AWS GovCloud (US-East) have inference requests processed via Amazon Bedrock in AWS GovCloud (US-West), while content remains stored in the region where the Kiro profile was created; all cross-region traffic is encrypted with TLS 1.2+.
  • MCP integration: Native Model Context Protocol support allows Kiro to connect to documentation, databases, APIs, and other enterprise resources to support complex development workflows.
  • Data isolation: Content collection for service improvement (prompts, responses, generated code) is disabled in GovCloud; user activity metrics and S3 reporting are also unavailable, and enterprise admins cannot enable telemetry.
  • Authentication: Social provider and AWS Builder ID logins are not available; authentication relies on IAM and SSO mechanisms appropriate for GovCloud environments.
  • Hooks and automation: Developers can automate repetitive workflows via event-triggered hooks that generate documentation and unit tests in the background.

Why It's Important

  • Unlocks AI-assisted development for the most sensitive workloads: FedRAMP High and DoD IL-4/5 are the highest civilian and defense compliance tiers, meaning agencies handling Controlled Unclassified Information (CUI) and even some classified data can now use agentic AI coding tools legally and safely.
  • Reduces the compliance barrier for AI adoption in government: Previously, federal developers had limited access to modern AI coding assistants due to data residency and authorization requirements; Kiro's authorization removes that blocker.
  • Accelerates mission-critical software delivery: Government development teams can leverage spec-driven automation to move faster from requirements to production-ready code, directly impacting program timelines.
  • Maintains data sovereignty: All content stays within AWS GovCloud (US) boundaries, with telemetry and data-sharing features disabled, giving agencies full control over sensitive information.
  • Signals broader AWS commitment to GovCloud AI parity: Kiro's rapid authorization demonstrates AWS's intent to bring cutting-edge AI/ML developer tooling to regulated sectors at pace with commercial availability.

How It's Different

  • Spec-driven vs. prompt-driven: Unlike most AI coding assistants that respond to ad hoc prompts, Kiro produces structured specifications, architectural plans, and task sequences before writing code, reducing ambiguity and rework.
  • Property-based testing: Kiro uses automated reasoning and property-based (fuzz-style) tests to validate code correctness beyond what unit tests cover, catching edge cases that pass standard test suites but fail in production.
  • Parallel agents: Kiro can deploy multiple agents working concurrently across a large codebase, enabling faster implementation of complex features compared to single-threaded AI assistants.
  • GovCloud-native with deliberate feature scoping: Rather than a simple lift-and-shift, Kiro's GovCloud variant explicitly disables features (telemetry, social login, inline suggestions) that could create data leakage risks, making the compliance posture transparent and auditable.
  • Enterprise governance built-in: IAM/SSO authentication, usage dashboards, cost controls, IP indemnity, and administration controls are first-class features, not afterthoughts, distinguishing it from developer-focused tools that lack enterprise governance.
  • Open standards compatibility: Kiro supports Agent Client Protocol (ACP), MCP, AGENTS.md, Open VSX extensions, and standard shells, avoiding vendor lock-in compared to proprietary AI IDE ecosystems.

When to Prefer It

  • Federal civilian agencies handling CUI or high-impact systems: Any agency operating under FedRAMP High requirements for systems processing sensitive but unclassified data should consider Kiro as their compliant AI development platform.
  • DoD programs at IL-4 or IL-5: Defense contractors and military program offices developing software for systems that process Controlled Unclassified Information or mission-critical DoD data can use Kiro without additional authorization overhead.
  • Large-scale government software modernization projects: When migrating legacy systems or building new platforms with complex requirements, Kiro's spec-driven approach helps ensure requirements traceability and architectural consistency.
  • Teams requiring strict data residency and no telemetry: Organizations that cannot allow any code, prompts, or responses to leave their authorized boundary will benefit from Kiro's disabled data-collection posture in GovCloud.
  • CI/CD pipelines in classified or sensitive environments: The CLI mode enables headless, automated use in pipelines where an interactive IDE is impractical, such as automated code review or bug-fix workflows in secure build environments.
  • Development teams needing MCP-connected enterprise context: When developers need to pull context from internal documentation, databases, or APIs during code generation, Kiro's native MCP support provides that integration within the GovCloud boundary.

Availability

  • Authorization status: Generally Available (GA) with FedRAMP High and DoD CC SRG IL-4 and IL-5 authorizations as of June 25, 2026.
  • Supported regions: AWS GovCloud (US-West) (us-gov-west-1) and AWS GovCloud (US-East) (us-gov-east-1); inference for US-East customers is processed in US-West via CRIS.
  • Default model: Claude Sonnet 4.5; automated model selection (Auto) is disabled at launch in GovCloud.
  • Pricing: Not explicitly stated in the announcement; contact your AWS account team for GovCloud-specific pricing details.
  • Feature limitations in GovCloud: VS Code plugin, inline suggestions, Autonomous Agent mode, social/Builder ID login, telemetry, user activity metrics, S3 reporting, and automated model selection are all unavailable.
  • Access method: Standalone IDE or CLI only; no VS Code plugin integration in GovCloud.
  • Documentation: Available at the AWS GovCloud User Guide (https://docs.aws.amazon.com/govcloud-us/latest/UserGuide/govcloud-kiro.html) and the Kiro product page (https://kiro.dev/).

Tags

Serviceskiro
Typesecurityregion-expansion
Conceptsagentic-aicoding-assistant
Use Casesgovernmentdeveloper-tools
GeographyAMERICAS

Related Resources

AI Radar AWS

AWS AI/ML news — curated, researched, explained

An automated intelligence platform that curates, researches, and analyzes AWS AI/ML/GenAI announcements daily. Every report is backed by real research — the system reads linked blog posts and documentation to provide accurate, in-depth analysis.

How Each Report Is Generated

  1. Collection — Daily monitoring of the AWS "What's New" RSS feed
  2. Filtering — AI-powered relevance detection for AI/ML/GenAI topics
  3. Taxonomy Tagging — LLM-based classification across 6 dimensions
  4. Importance Scoring — Point-based system with tag bonuses (1-5 stars)
  5. Research Phase — Follows links to blog posts and documentation
  6. Report Generation — Claude Sonnet produces structured 6-section analysis
  7. Visual Summary — Claude Opus generates Mermaid diagrams for key items
  8. Publishing — Static website rebuilt and deployed via CloudFront

Features

  • Faceted filtering by service, type, concept, and more
  • Multi-dimensional taxonomy with 80+ tags across 6 dimensions
  • Geographic availability badges (Global, APJ, EMEA, AMER) with filtering
  • Timeline visualization of announcement volume
  • PDF export for offline reading
  • Mermaid visual summaries for key announcements
  • Daily automated updates — no manual curation
What makes this different: Each report involves a dedicated research phase where the system reads linked blog posts and AWS documentation pages. This produces analysis that goes beyond the original announcement text.

Technology

Built with Python, AWS Lambda, Amazon Bedrock (Claude Sonnet 4.6, Opus 4.6, Haiku 4.5), S3, CloudFront, WAF, EventBridge, and CDK.

Open Source

This project is open source. Fork it, customize it for your needs, and deploy your own instance.
📦 github.com/bbonik/ai-radar-aws

How Importance Scoring Works

Each announcement receives a point score based on multiple factors. The total score maps to a 1-5 star rating:

1★ < 2 pts 2★ ≥ 2 pts 3★ ≥ 3.5 pts 4★ ≥ 5 pts 5★ ≥ 6.5 pts

Point Breakdown

FactorPointsWhen
Core AI service (Bedrock, AgentCore, SageMaker AI)+4Service named in title
Key AI service (SageMaker, Kiro, QuickSight)+2Service named in title
Other AI-related service+1Default
Blog post link+3Link to aws.amazon.com/blogs/
GitHub samples link+2Link to github.com/aws*
Documentation link+1Link to docs.aws.amazon.com/
New model+1.5Tagged as "new-model"
New service+1Tagged as "new-service"
New feature+0.5Tagged as "new-feature"
Anthropic / OpenAI provider+2Provider explicitly mentioned
Instance / notebook announcement-2Hardware/capacity, not feature
Performance / pricing / security-0.5Incremental updates
Region expansion to APJ+1Expands to Asia Pacific
Region expansion (non-APJ only)-1.5Only expands to other regions

Geographic Relevance Badges

Each announcement card shows a small badge indicating whether the feature is available in your region:

🌐 Global Available in all regions
🌏 APJ Asia Pacific
🌍 EMEA Europe / Middle East / Africa
🌎 AMER Americas (US, Canada, South America)
No badge Geography unknown
How geography is detected: The system detects ALL geographies mentioned in each announcement. If the text mentions specific regions (Tokyo, Frankfurt, Oregon, etc.), the corresponding geography badges are shown. If it says "all regions" or is a new feature with no region specified, it gets the Global badge. Geography is also filterable — click a geo chip to see only announcements available in that region.