Agent skills for Claude Desktop

Use when designing multi-agent orchestration — patterns for delegation and coordination.

Use when applying web best practices — security headers, modern APIs, error handling.

Use when setting up or modifying build and deploy pipelines — shift left, quality gates.

Use when output quality drops or you switch tasks — feed agents the right information at the right time.

Use when diagnosing or optimizing LCP, INP, CLS — Core Web Vitals at the 75th percentile (LCP ≤2.5s, INP ≤200ms, CLS ≤0.1).

Use when implementing changes touching multiple files — thin vertical slices, feature flags, rollback-friendly.

Use when the request is underspecified — interview one question at a time until 95% confidence.

Use when building an MCP server from an OpenAPI spec.

Use when optimizing web performance — field/lab measurement, traces, and loading fixes.

Use when optimizing performance — measure first, then profile, bundle, and fix loading or runtime.

Use when you have a spec and need small verifiable tasks with acceptance criteria.

Use when acting as product manager — RICE, PRDs, backlog, and roadmap.

Use when tracking SaaS metrics — MRR, churn, CAC, and financial analysis.

Use when handling user input, auth, secrets, or external integrations — prevent OWASP Top 10, enforce auth patterns, secrets management…

Use when you need authoritative framework code — verify and cite official docs.

Use when starting a new project or significant change — write a PRD before any code.

Use when assessing tech debt — scan, score severity, and generate prioritized remediation plans.

Use when auditing web quality — orchestrate performance, a11y, SEO, best-practices across Lighthouse categories.

Azure Functions, Cosmos DB modeling, Service Bus patterns, Bicep templates

Self-contained parallel generator — invoke directly, do not decompose.

Agent-native headless browser. 26 primitives over a Unix-socket wire protocol.