FAI Skills
Reusable capability modules — each a LEGO block. Search the catalog, inspect evidence and token size, compare Full and Lean variants, then copy, download, or install from the same bounded record.
Discover
Semantic and voice search with domain filters
Inspect
Quality, tokens, WAF, events, and Lean evidence
Install
CLI and native VS Code deep links
Evidence catalog
Find the right skill for the job.
Filter by operating concern, search by intent, and inspect the source contract before using it.
886,378 Full tokens · switch to Lean for exact per-card and filtered-view receipts
638/638 gatedLean Mode638 skills · one-click copy · installable in 70+ agents · Claude Code marketplace · exact token sizes + Quality score.
npx skills add frootai/frootaiOrchestrate builder, reviewer, and tuner agent handoffs for solution plays — define roles, model routing, token budgets, and quality gates
Configure API rate limiting, APIM throttling, and retry behavior for AI endpoints — absorb bursts, isolate noisy tenants, and control 429 retries
Configure Application Insights, OpenTelemetry, and KQL dashboards for AI workloads — trace latency, token usage, failures, and cost signals
Implement immutable audit logging, PII redaction, and Sentinel monitoring for AI systems — preserve evidence, prove access history, and detect suspicious activity
Set up Azure OpenAI deployments, RBAC, and monitoring — harden inference endpoints and control latency, quota, and cost
Back up AI data, vector indexes, and configuration state — recover prompts, conversations, and search assets after failures
Create reusable Bicep modules, parameter files, and registry packages — standardize Azure infrastructure and reduce deployment drift
Implement canary deployment, rollout gates, and rollback automation for AI changes — catch latency or quality regressions before full release
Add a circuit breaker, fallback routing, and health checks for Azure OpenAI calls — contain 429 and 503 cascades and preserve service
Run AI compliance audits, collect Azure evidence, and track remediation — assess SOC 2, HIPAA, ISO 27001, and EU AI Act controls
Select embedding models, dimensions, and hybrid search settings for vector workloads - balance recall, storage cost, and migration risk
Configure builder→reviewer→tuner agent chain with handoff rules, model assignment, and evaluation gates.
Review agent safety controls, budget limits, and human escalation paths - approve only agents that stay bounded, observable, and policy-compliant
Run evaluator-optimizer loops for agents and prompts - score outputs, revise weak instructions, and stop only when quality thresholds hold
Generate API reference docs, examples, and error catalogs from OpenAPI or code annotations - keep endpoint docs accurate, testable, and publishable
Generate typed REST API endpoints with Zod/Pydantic input validation, RFC 7807 error responses, OpenAPI annotation, and Managed Identity auth — eliminating schema drift and IDOR vulnerabilities.
Configure API rate limiting with Azure APIM policies, throttling tiers, and retry headers.
Configure Application Insights with custom AI metrics, anomaly alerts, and operational dashboards.
Generate solution architecture blueprints with Mermaid service topology, component responsibility tables, data flow annotations, and WAF pillar mapping — bridging system design to implementation.
Generate Architecture Decision Records capturing context, forces, decision rationale, rejected alternatives, and consequences — preventing knowledge rot when teams diverge from original design intent.
Wire .NET Aspire AppHost for multi-container orchestration, automatic service discovery, distributed OpenTelemetry export, and local Azure emulator integration — eliminating hardcoded connection strings and manual container start order.
Scaffold ASP.NET Core Minimal API with typed route groups, FluentValidation, Managed Identity credentials, OpenAPI v3 via Scalar, Serilog structured logging, and health check endpoints — cutting AI service setup time from hours to minutes.
Implement immutable audit logging with Azure Blob Storage, Sentinel integration, and compliance evidence.
Analyze Azure AI workload spend — Cost Management API queries, overprovisioned PTU flags, semantic-cache gaps, model-SKU right-sizing, and FinOps recommendations with ROI estimates. USE FOR: reducing Azure AI/OpenAI bills, finding cost anomalies, deciding PTU vs pay-as-you-go, right-sizing deployments, building a FinOps report. DO NOT USE FOR: non-Azure cloud costs, generic budgeting, infrastructure provisioning, performance tuning unrelated to cost. Triggers: "azure cost", "finops", "reduce spend", "ptu", "cost optimization", "overprovisioned", "save money", "bill", "budget".
Provision Azure AI Foundry Hub and Project with Managed Identity RBAC, private endpoint networking, connected Azure AI Services and Key Vault — ready for Prompt Flow evaluation pipelines and model fine-tuning.
Create Azure AI Search vector indexes — HNSW profiles, semantic ranker, hybrid BM25+vector search, field mappings for chunked RAG documents, and batch indexing pipelines. USE FOR: building a RAG retrieval index, fixing poor recall or irrelevant chunks, configuring hybrid/semantic search, designing index schema, batch ingestion of documents. DO NOT USE FOR: Cosmos DB modeling, embedding-model selection (use the embedding skill), non-Azure search engines (Elastic, Pinecone), full-text-only search without vectors. Triggers: "ai search", "vector index", "hnsw", "semantic ranker", "hybrid search", "rag index", "retrieval", "chunk", "recall".
Wire Azure App Configuration with feature flag targeting filters, Key Vault reference chaining, .NET/Python SDK refresh polling, and blue/green rollout strategies — eliminating hard-coded config values and restart-required deployments.
Conduct Azure Well-Architected Framework reviews with scored pillar assessments, tiered finding severity, remediation priority tables, and an executive scorecard — turning ad-hoc architecture debates into structured, trackable decisions.
Design and apply Azure Blob Storage lifecycle management policies with tiering to cool/archive, retention rules, and delete-after-N-days automation — reducing storage costs by 70% without manual archival work.
Integrate Language, Speech, Vision, and Translator services with Managed Identity authentication, pre-built and custom models, content safety filters, and regional routing — enabling multi-modal AI without managing credentials.
Configure Azure Container Registry with geo-replication across regions, image scanning for vulnerabilities, Managed Identity pull access, and OCI artifact support — enabling secure, distributed container delivery.
Design Cosmos DB data models — partition keys, Request Unit (RU) estimation, vector-search embedding storage, and analytical workload isolation for RAG and time-series workloads. USE FOR: choosing a partition key, fixing hot partitions or RU spikes, modeling chat/conversation history, storing vector embeddings, isolating analytics from OLTP, denormalization and indexing strategy. DO NOT USE FOR: relational schema design (use Azure SQL), graph modeling, Cosmos infrastructure provisioning (use the Bicep skill), non-Azure databases. Triggers: "cosmos db", "partition key", "RU", "request units", "hot partition", "vector store", "nosql model", "data model", "embedding storage".
Configure Azure Data Explorer for real-time telemetry ingestion with KQL queries, time-series aggregations, and interactive dashboards — analyzing AI workload metrics and traces at petabyte scale.
Configure Azure Event Grid for event-driven AI pipelines with topic subscriptions, dead-letter queues, retry policies, and role-based filtering — routing document uploads to embeddings to RAG queries without polling.
Configure Azure Event Hubs for real-time data ingestion with stream processing, auto-scaling, Managed Identity auth, and consumer groups — enabling AI workload telemetry at millions of events/second.
Scaffold Azure Functions apps with HTTP/Event Hub/Service Bus triggers, input/output bindings, Managed Identity, Application Insights, and local development setup — enabling serverless AI microservices in minutes.
Configure Azure Key Vault with managed identities, RBAC, secret rotation policies, customer-managed keys (CMK), and app integration patterns — avoiding hardcoded secrets and enabling compliant AI workload credential management.
Deploy Azure OpenAI with model selection, PTU vs PAYG pricing tiers, content safety filters, token management strategies, and streaming patterns — maximizing cost efficiency and compliance for AI applications.
Set up Azure OpenAI Service with model deployments, RBAC, content filtering, and monitoring.
Query Azure Resource Graph across subscriptions with KQL for compliance audits, cost analysis, RBAC reviews, and cross-tenant resource discovery — enabling unified visibility into sprawling AI infrastructure.
Diagnose Azure resource health issues via Resource Health events, platform metrics, activity logs, and one-click remediation workflows — preventing cascading AI infrastructure failures.
Generate visual topology diagrams of Azure resources with dependency overlays, compliance annotations, and cost heatmaps. Use when creating architecture documentation, reviewing resource layouts, or presenting infrastructure to stakeholders.
Select least-privilege Azure RBAC roles by mapping required actions to built-in roles. Use when assigning permissions to Managed Identities, service principals, or users for AI workloads.
Configure Azure Service Bus queues and topics with dead-letter handling, sessions, retry policies, and throughput tuning. Use when building reliable messaging for AI pipeline orchestration, async processing, or event-driven architectures.
Provision Azure SQL Database with AAD-only auth, private networking, backup policies, and workload performance tuning. Use when setting up relational storage for AI application metadata, conversation history, or evaluation results.
Set up Azure Static Web Apps with custom auth, staged preview environments, API backend linking, and route-level access control. Use when deploying frontend apps with serverless APIs, GitHub/Azure DevOps CI/CD, and role-based route protection.
Apply Azure Storage design patterns for blobs, queues, tables, and files with lifecycle management, private access, SAS policies, and replication strategies. Use when designing storage architecture for AI data pipelines.
Set up backup and restore procedures for AI application data, model configs, and index state.
Optimize baseline prompts with role framing, output constraints, few-shot examples, and deterministic output validation. Use when improving prompt quality, reducing hallucinations, or standardizing LLM output format.
Create a reusable Bicep module with typed parameters, conditions, outputs, and AVM alignment.