Soham
Community@sohamda · Netherlands
Soham's registry delivers 49 MIT-licensed agent skills for Azure deployment, infrastructure-as-code, AI model operations, governance, and architecture documentation.
Agent Skills by Soham
Showing 49 vetted skills indexed across 1 GitHub repositories.
azure-diagrams
Routes diagram generation requests to specialized drawio, python-diagrams, or mermaid skills.
microsoft-foundry
Deploy, evaluate, and manage Azure AI Foundry agents and model deployments end-to-end.
azure-hosted-copilot-sdk
Scaffold and deploy GitHub Copilot SDK applications to Azure Container Apps.
appinsights-instrumentation
Provides guidance and SDK references for instrumenting webapps with Azure Application Insights telemetry.
copilot-customization
Guides selection and authoring of VS Code Copilot customization mechanisms including instructions, prompts, agents, skills, and MCP servers.
count-registry
Provides canonical entity counts from count-manifest.json to prevent hard-coded numbers in documentation.
microsoft-code-reference
Verify Microsoft SDK code against official documentation and find working code samples.
azure-aigateway
Configure Azure API Management as an AI Gateway for models, MCP tools, and agents.
azure-ai
Implements Azure AI Search, Speech, OpenAI, and Document Intelligence integrations via MCP tools and SDKs.
make-skill-template
Scaffolds new Agent Skills with SKILL.md frontmatter, folder structure, and bundled resources.
azure-messaging
Diagnose and resolve Azure Event Hubs and Service Bus SDK connection, authentication, and message processing issues.
microsoft-skill-creator
Generate agent skills for Microsoft technologies using Learn MCP documentation tools.
deploy-model
Deploys Azure OpenAI models with intent-based routing across preset, customize, and capacity discovery modes.
preset
Deploys Azure OpenAI models to regions with available capacity using Azure CLI.
capacity
Discovers Azure OpenAI model capacity across regions and ranks deployment locations by quota and projects.
customize
Deploys Azure OpenAI models through an interactive guided workflow with full control over version, SKU, capacity, and content filtering.
golden-principles
Defines ten operating principles governing agent behavior in repository-driven infrastructure workflows.
workflow-engine
Routes multi-step agent workflows through a machine-readable DAG with gates and fan-out.
vendor-prompting
Audit agent and prompt files against Anthropic Claude and OpenAI GPT-5.5 prompting best practices.
azure-diagnostics
Diagnose and troubleshoot Azure Container Apps and Function Apps using KQL queries and health checks.
terraform-search-import
Discovers existing Azure resources and generates Terraform import blocks for bulk adoption into IaC.
azure-compliance
Audits Azure resources for compliance issues and Key Vault expiration risks using azqr scans.
terraform-test
Write and run Terraform .tftest.hcl tests for Azure infrastructure modules.
azure-governance-discovery
Discovers effective Azure Policy assignments and emits schema-compliant governance-constraints JSON.
Frequently Asked Questions About Soham
FAQPage SchemaWhat tasks can I perform using Soham's Azure skills?▼
You can scaffold and deploy Azure apps with Bicep or Terraform, deploy Azure OpenAI models with capacity discovery, assign least-privilege RBAC roles, run compliance and cost audits, troubleshoot Container Apps and Event Hubs, query Kusto with KQL, and generate Draw.io, Mermaid, or matplotlib architecture diagrams.
Who are these skills designed for?▼
Cloud engineers, platform teams, and solution architects working in Microsoft Azure. The skills target practitioners handling IaC authoring, pre-deployment validation, AI model provisioning, governance discovery via Azure Policy, and production diagnostics across subscriptions.
How do the skills run in practice?▼
Skills load as SKILL.md files into VS Code Copilot, Claude Code, or any MCP-compatible host. Several invoke MCP servers such as Microsoft Learn, drawio, azure-quota, and azure-kusto; deployment skills chain azure-prepare, azure-validate, and azure-deploy with azd or terraform apply.
Are these skills free and open source?▼
Yes. Nearly all skills carry an MIT license, including the Microsoft-authored deployment, diagnostics, and AI gateway skills. A few utility skills such as count-registry, iac-common, and golden-principles list no explicit license in their frontmatter.
What prerequisites do the skills require?▼
Requirements vary: Azure CLI for gateway and quota operations, Terraform 1.6-1.14 with azurerm ~> 4.0 for testing and import skills, Python 3.14 for governance discovery and context auditing, Bicep extension for Bicep patterns, and read access to target Azure subscriptions.