zxkane/aws-skills

This system is a plugin and skill management framework for AWS infrastructure and AI agent development. It provides structured guidance, templates, and validation tools for AWS CDK, SST, and AgentCore deployments. The system enforces marketplace integrity and configures AI assistant skills to support AWS-specific workflows.

Added six new template files in the scripts directory to support deploying agents on AWS AgentCore. This includes a Dockerfile for ARM64 runtime builds, Python server templates for A2A, AG-UI, and MCP protocols, a FastAPI-based HTTP runtime template with SSE streaming, and a CDK custom resource Lambda for managing AgentCore Gateway lifecycle. These templates provide the foundational code for users to deploy and manage AI agents on the platform.

plugins/aws-agentic-ai/skills/aws-agentic-ai/scripts · high confidence

A new Python script, scripts/ci/validate\_marketplace.py, has been added to validate the plugin marketplace structure. It checks that marketplace.json is valid, verifies that each plugin's source directory exists, ensures every listed skill resolves to a directory containing a SKILL.md file, confirms that SKILL.md files have the required YAML frontmatter (name and description), validates that symlinks in .claude/skills/ are not dangling or pointing outside the repository, and detects orphaned skills not registered in the marketplace. This script is designed to be run from the repository root and will fail the build if any of these conditions are not met.

scripts · high confidence

The plugins/aws-iac area now includes new skill definitions for AWS infrastructure-as-code workflows. A new 'aws-cdk-development' skill has been added to guide users through AWS Cloud Development Kit (CDK) patterns, naming conventions, and validation scripts. Additionally, a new 'aws-sst-development' skill has been introduced for AWS SST v4 (Ion) development, covering authoring, testing, and deployment references for Pulumi-backed stacks. These additions provide structured guidance and tooling for managing AWS infrastructure using CDK and SST frameworks.

plugins/aws-iac · high confidence

The .claude directory now includes a settings.json file that enables several official plugins (skill-creator, plugin-dev, pr-review-toolkit, code-review, security-guidance, and security-scanning). Additionally, symlinks were added in .claude/skills/ to point to various AWS-related skills (aws-agentic-ai, aws-cdk-development, aws-cost-operations, aws-mcp-setup, aws-serverless-eda, and aws-sst-development) located in the plugins directory, making these skills available for use.

.claude · high confidence

Added comprehensive documentation for the new AgentCore services: Browser, Code Interpreter, Evaluations, Gateway, Identity, Memory, and Observability. Each service includes an overview, core capabilities, use cases, architecture diagrams, configuration examples, and best practices. The Gateway service also includes deployment and validation scripts.

plugins/aws-agentic-ai/skills/aws-agentic-ai/services · high confidence

Written by watchdog.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.

• CAI 50 → 51 (+0.7)
• Rubric changed (rubric-2026.08.19 → rubric-2026.09.15) — scores are not directly comparable.

• Code Health 100 → 100 (+0.0)
• Architecture 69 → 69 (+0.0)
• Maturity 59 → 59 (+0.0)
• Readiness 29 → 29 (+0.0)
• Security 90 → 99 (+9.1)

• Dependency hygiene not measured — no supported dependency manifest was read
• High: security finding (details withheld)
• High: security finding (details withheld)
• LLM evaluation failed
• Medium IaC: CKV_DOCKER_2 (plugins/aws-agentic-ai/skills/aws-agentic-ai/scripts/Dockerfile.runtime-template)
• No exposed public API
• early-stage repository — too few commits for a meaningful bus factor
• early-stage repository — too little history to judge knowledge freshness

• Documentation: no architecture or design documentation (docs/aws-mcp-setup.md)
• Documentation: no installation or build instructions (plugins/aws-agentic-ai/skills/aws-agentic-ai/services/browser/README.md)
• Documentation: no installation or build instructions (plugins/aws-agentic-ai/skills/aws-agentic-ai/services/code-interpreter/README.md)
• Documentation: no project overview (README.md)
• High: security finding (details withheld)
• High: security finding (details withheld)
• Medium IaC: WD-DOCKER-0003 (plugins/aws-agentic-ai/skills/aws-agentic-ai/scripts/Dockerfile.runtime-template)
• Medium IaC: WD-DOCKER-0003 (plugins/aws-agentic-ai/skills/aws-agentic-ai/scripts/Dockerfile.runtime-template)
• Medium IaC: WD-DOCKER-0010 (plugins/aws-agentic-ai/skills/aws-agentic-ai/scripts/Dockerfile.runtime-template)

Written by watchdog.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.

zxkane/aws-skills was measured the same way every project in this corpus was: the same rubric, at a pinned commit, with the result published in full. Point a surveyor at a repository you know and see whether you agree with it.