mrsknetwork/supernova Skills

Integrates AI language model capabilities into applications — OpenAI (GPT-4o), Anthropic (Claude), and Google Gemini APIs, including streaming responses, function calling/tool use, retrieval-augmented generation (RAG) with vector search, and prompt management. Use when adding any AI features: chatbots, document Q&A, content generation, classification, summarization, or agents. Trigger when user mentions "OpenAI", "Claude", "Anthropic", "GPT", "AI chat", "chatbot", "RAG", "embeddings", "vector search", "LLM", "AI-powered", or "generate content".

Designs and implements HTTP APIs in Python using FastAPI (REST) or Strawberry (GraphQL). Enforces contract-first design, standard response envelopes, authentication dependency injection, and rate limiting. Trigger when building API endpoints, designing schemas, or handling client-server communication. Always choose REST vs GraphQL based on application complexity.

Performs structured code, security, and architecture audits. Produces severity-categorized findings with file/line evidence and actionable remediation steps. Use when reviewing a PR, conducting a security review, evaluating technical debt, or assessing code quality before a release. Never merge critical audit findings without a documented resolution.

Integrates third-party authentication — OAuth social login (Google, GitHub) and managed auth providers (Clerk, Auth0, Supabase Auth). Use when a user wants to add "Sign in with Google/GitHub", set up single sign-on, or delegate authentication to a managed service instead of rolling their own JWT system. Trigger when user mentions "OAuth", "social login", "Google login", "GitHub login", "Clerk", "Auth0", "Supabase Auth", "SSO", or "I don't want to build my own auth". Routes to the `security` skill if the user wants custom JWT auth from scratch.

Implements server-side application logic, data access layers, and background services using Python 3.12 and FastAPI. Use for building service classes, repositories, background tasks, middleware, or any Python code that runs on the server. Always ask about the existing stack before applying defaults.

Designs and implements the core domain rules, service-layer workflows, state machines, and validation logic that define what an application actually does. Use when implementing rules that go beyond simple CRUD, modeling complex workflows (order processing, subscription billing, approval chains), or translating product requirements into enforceable system behavior. Trigger when a feature involves conditional rules, multi-step processes, state transitions, or calculations that have business meaning.

Designs relational database schemas, writes optimized SQL queries, manages Alembic migrations, and configures RxDB for offline-first client-side applications. Use for all data modeling, schema design, query optimization, and migration tasks. Always confirm the existing stack before applying PostgreSQL or RxDB defaults.

Designs and implements CI/CD pipelines, Docker containerization, multi-environment deployment, and release strategies. Use when setting up automated builds, deploying an application for the first time, adding tests to a pipeline, or managing release processes. Always confirm the existing hosting provider and repository before applying templates.

Guide users through a structured workflow for co-authoring pristine, standardized documentation (READMEs, proposals, specs). Focuses on clarity, formatting, and reader comprehension. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks.

Implements transactional email via Resend or SendGrid — welcome emails, password resets, order confirmations, and notification digests. Covers HTML template rendering, queue-based sending via Celery, unsubscribe handling, and bounce management. Use when adding any email to an application. Trigger when user mentions "send email", "email notification", "welcome email", "password reset email", "Resend", "SendGrid", "SMTP", "transactional email", or "email template".

Strictly and autonomously implements a single task item from a planner-generated checklist. Enforces the rule of exploring before writing, applying atomic edits, running self-validation, and reporting status. Use this skill when a specific task item is ready to be implemented. Always use executor for individual checklist items rather than trying to implement everything in one shot.

Implements file upload, storage, and retrieval using AWS S3 or Cloudflare R2, including pre-signed URLs, image resizing with Pillow/sharp, CDN delivery, and secure MIME validation. Use when users need to upload files — profile pictures, document attachments, product images, or any user-generated media. Trigger when user mentions "file upload", "image upload", "S3", "R2", "Cloudflare", "CDN", "pre-signed URL", "avatar", "attachments", or "store files". Never stores uploaded files on the local filesystem in production.

Implements client-side user interfaces using Next.js 14+ App Router, TypeScript strict mode, Tailwind CSS v3, Shadcn/ui, and Radix UI. Enforces Server vs Client component boundaries, typed props, accessible markup, and integrated form handling. Use for all React/Next.js components, page layouts, data fetching, and styling tasks. Always ask about the existing stack before applying defaults.

Provisions and manages cloud infrastructure using Terraform, configures Kubernetes deployments, and manages network and IAM resources. Use when provisioning cloud environments, setting up Kubernetes clusters, writing Terraform modules, or configuring network topology. Always confirm the cloud provider and existing infra state before making changes.

Manages database schema changes using Alembic — generates migration files, reviews them for safety, runs them in the right order, and handles rollbacks. Use when adding a new table, modifying columns, adding indexes, or renaming anything in the database. Trigger when the user mentions "migration", "alembic", "schema change", "add column", "rename table", "database version", or "upgrade/downgrade database". This skill prevents the most common data-loss mistakes: running migrations in production without a backup, generating empty migrations, and forgetting to handle nullability on existing rows.

Sets up production observability — error tracking with Sentry, structured logging with structlog, application metrics with Prometheus, and uptime monitoring. Use when preparing an app for production, debugging production issues, or when the user has no visibility into what's failing in their live app. Trigger when user mentions "Sentry", "error tracking", "monitoring", "logging", "metrics", "alerts", "production issues", "how do I know when something breaks", or "observability".

Sets up a new project from scratch — creates the complete repository structure, configuration files, environment setup, and base scaffolding so development can start immediately. Use this as the FIRST skill on any brand-new project before any other skill runs. Trigger when the user says "start a new project", "scaffold a new app", "I want to build X from scratch", "initialize project", or when no code exists yet and the user is describing an idea.

The entry-point coordinator for complex multi-domain builds. Classifies user intent, selects appropriate domain skills, enforces phase gates, and manages the overall build workflow. Use this skill whenever a user's request spans multiple technical domains (e.g., "build a SaaS app") or when coordinating output from multiple other skills. Always route through orchestrator before executing any domain work.

Coordinates simultaneous execution of independent tasks to reduce total build time. Use this skill when multiple tasks from the plan checklist have no dependencies between them and can be safely executed concurrently. Trigger when the plan has tasks in the same phase that do not share data or files.