ravn-ai-toolkit

Add a new rule, convention, or instruction to the project's agent configuration. Analyzes the rule and helps decide placement: root CLAUDE.md (universal rules), docs/agents/ files (topic-specific guidance), or a new skill (complex workflows). Use when users say: '/agent-add-rule', 'add a rule', 'add convention', 'new coding standard', 'add instruction for claude', 'update claude.md with'.

Initialize or migrate to nested CLAUDE.md structure for progressive disclosure. Claude auto-loads CLAUDE.md from any directory it enters, so nested files get discovered automatically. Use when setting up a new project's agent config, refactoring a bloated CLAUDE.md, or adding progressive disclosure to an existing repo. Triggers on: '/agent-init-deep', 'setup progressive disclosure', 'refactor claude.md', 'split claude.md', 'claude.md is too big'.

Analyzes git diffs and commit history to intelligently fill PR templates and create pull requests via gh CLI. Use when: creating a PR, needing PR description help, saying 'create a pull request', 'fill PR template', 'make a PR', 'open a pull request'. Triggers on: create PR, open pull request, make a PR, fill template.

Guide for creating effective, portable skills that extend Claude's capabilities with specialized knowledge, workflows, and tool integrations. Use when: (1) Creating a new skill from scratch, (2) Updating or improving an existing skill, (3) Structuring skill content for progressive disclosure, (4) Writing skill descriptions and triggers, (5) Packaging a skill for distribution. Triggers on: "create a skill", "build a skill", "new skill", "update this skill", "improve skill description", "skill structure", "skill triggers".

Manage AI skills from the Ravn AI Toolkit via corvus CLI — install, update, remove, search, and configure skills for any project. Use when: (1) Installing AI skills into a project, (2) Updating installed skills to latest versions, (3) Browsing or searching available skills, (4) Configuring global or per-project skill sets, (5) Troubleshooting corvus setup. Triggers on: "install skills", "add skills", "update skills", "corvus", "skill manager", "browse skills", "set up AI rules".

Behavioral compliance testing for any CLAUDE.md or agent definition file. Auto-generates test scenarios from your rules, runs them via LLM-as-judge scoring, and reports a compliance score with per-rule pass/fail breakdown. Optionally improves failing rules via automated mutation loop. Use when: (1) testing whether your CLAUDE.md rules are actually followed, (2) evaluating an agent definition for role-boundary compliance, (3) dogfooding a skill's own SKILL.md. Triggers on: "eval", "compliance test", "test my CLAUDE.md", "check rules", "behavioral test", "/eval-agent-md". Do not trigger for: editing or writing CLAUDE.md rules, general code review, adding linting config, or any task that is not explicitly about testing behavioral compliance.

Interview the user relentlessly about a plan or design until reaching confidence through Socratic questioning. Use when: stress-testing a plan, probing design assumptions, validating feasibility. Triggers on: grill me, stress-test, interview me, validate plan, probe design.

Run a task in a background sub-agent so you can continue working on other things. Use when explicitly invoked via the '/parallel' slash command. Never auto-triggers from natural language.

Reads open review comments from a GitHub PR, triages them by severity, applies code fixes, and drafts replies. Use when: addressing PR comments, fixing review feedback, responding to code review. Triggers on: address review comments, fix PR feedback, handle PR comments, respond to review.

Transform user requests into detailed, precise prompts for AI models. Use when users say 'promptify', 'promptify this', 'rewrite this prompt', 'make this prompt better/more specific', or explicitly request prompt engineering or improvement of their request for better AI responses.

Rewrite a feature branch's commit history into clean conventional commits that tell a progressive, linear story. Handles backup, soft reset, and atomic recommit. Use when: (1) Cleaning up messy WIP commits before PR, (2) Reorganizing commits into logical units, (3) Converting commits to conventional commit format. Triggers on: "rewrite history", "clean up commits", "rewrite commits", "conventional commits", "squash and rewrite", "reorganize commits".

Convert meeting transcript .txt files into structured .md notes with metadata, TL;DR, key topics, action items, and quotes. Use when processing raw transcripts into formatted notes. Triggers on: "process transcript", "generate notes from transcript", "transcript to notes", "/transcript-notes".

Audit a repository for type-system weaknesses using recent bug-fix commits as hard evidence. Produces prioritized findings tied to specific commits showing which types allowed real bugs. Use when: reviewing type safety, auditing types, analyzing type bugs. Triggers on: type audit, type system review, audit types, type safety audit.

Server-side architecture and security — API design, error handling, validation, logging. Use when building APIs, server logic, or reviewing backend security.

tRPC end-to-end typesafe APIs. Router architecture, procedure design, middleware chaining, input validation with Zod, error handling, and Vertical Slice Architecture. Use when building tRPC APIs, designing procedures, structuring backend slices, or handling cross-procedure logic.

Design and implementation patterns for building command-line tools with modern UX. Use when designing a CLI, reviewing CLI UX, defining commands and flags, or implementing error handling and signal handling.

SQL database design, query optimization, and migration safety. Use when writing queries, designing schemas, or planning database migrations.

Drizzle ORM typesafe schema design, relational queries, prepared statements, migrations, and transactions. Use when working with Drizzle ORM, writing database queries, managing migrations, or optimizing query performance with prepared statements.

Generate a gallery of design variations for a UI component. Takes an existing component (referenced by name, pasted code, or screenshot) and produces N distinct rendered alternatives in a single comparison page. Use when exploring visual directions, generating mockups, comparing design approaches for a component, creating A/B candidates, or when anyone says "show me options" or "give me variations" for a UI element.

WCAG AA and ARIA best practices — screen readers, keyboard navigation, focus management. Use when building any user-facing interface or reviewing accessibility compliance.

Visual design system patterns for web UIs. Tailwind CSS v4 design tokens and CSS variables, responsive design with container queries, dark mode, layout patterns, and spacing scales. Use when implementing visual designs, working with Tailwind CSS, or building responsive layouts.

Convert Figma component designs into production-ready React implementations with design token integration, accessibility via React Aria, and comprehensive documentation. Use when building React components from Figma designs, generating component implementation specs, or bridging design-to-development workflows.

Framework-agnostic frontend architecture — state management, components, data fetching. Use when building any web frontend, choosing state patterns, or structuring UI code.

React 19 patterns for components, hooks, Server Components, and data fetching. Use when writing React components, managing state with hooks, implementing Suspense boundaries, optimizing renders with proper memoization, or building Server/Client component hierarchies.

Apple's Liquid Glass design system for iOS 26+ and iPadOS 26+. Use when building iOS 26+ UI with glassEffect, implementing GlassEffectContainer, working with glass morphing transitions, or migrating from UIKit to SwiftUI glass APIs.

Localizes UIKit Views and ViewControllers and SwiftUI Views. Extracts hardcoded text, generates camelCase keys, replaces literals with String(localized:) or LocalizedStringKey, and creates or updates Localizable.xcstrings. Use when: (1) localizing a UIKit or SwiftUI file, (2) extracting hardcoded strings from Swift, (3) adding xcstrings entries. Triggers on: localize, extract strings, add localization, make localizable, xcstrings, localize this view, localize this file. NOT for general Swift tasks unrelated to localization.

Swift Concurrency patterns — async/await, actors, tasks, Sendable conformance. Use when writing async/await code, implementing actors, working with structured concurrency, or ensuring data race safety.

Android and Kotlin development patterns — Compose, architecture, coroutines, Room, navigation, Hilt. Use when building Android apps, writing Jetpack Compose UI, or reviewing Android-specific code.

Draft, evaluate, enrich, and normalize QA bug reports to RAVN standards. Trigger on "write/create a bug report", "I found a bug", "log this defect", "evaluate/score/review/improve my bug report", "normalize these bug reports", or when a user describes unexpected behavior. Also triggered by /bugreport.

Scan a website URL and extract the best locators for every visible element on the page, outputting a Page Object Model (POM) code file and/or a JSON locator map tailored to a specific test automation framework (Playwright, Cypress, or WebdriverIO). Use when users say "scan this page for locators", "extract locators from URL", "get selectors from website", "generate page object from URL", "build POM from page", "find locators for this URL", "what locators should I use for this page", "scan this page for elements", "extract selectors", "generate locator map", "get element selectors for testing", or mention scanning/extracting locators from a website. Also trigger when a user pastes a URL and asks to generate a page object, locator map, or selectors for Playwright, Cypress, or WebdriverIO. Also explicitly triggered by the /locators command.

Receive bug reports from QA agents and implement focused, minimal fixes. Read before writing, understand the root cause, fix with the smallest possible change, and report what was changed. Works with any tech stack. Trigger on "fix the QA bugs", "implement the bug fix", "patch the defect", or when a QA agent has produced bug reports with reproduction steps.

Adversarial QA tester that systematically tries to break an application's API. Tests security boundaries, input validation, race conditions, deduplication, and malformed requests. Reports bugs with full reproduction details. Trigger on "break the API", "chaos monkey", "adversarial testing", "security test the endpoints", "test edge cases", or when a test plan defines API endpoints.

Drive a web application's UI through core positive flows using Playwright browser automation. Verify that primary user journeys work end-to-end, take snapshots at each step, and report bugs with full reproduction details. Trigger on "run happy path tests", "test the UI flows", "QA the happy path", "verify the user journey", or when a test plan defines UI flows.

Orchestrate QA agent workflows — spawn test agents in parallel, collect results, triage bugs, trigger the bug fixer, and generate QA reports. The main entry point for running a QA session. Trigger on "run QA", "start QA session", "test the PR", "orchestrate QA agents", or when the user wants to run multiple QA agents together. Also triggered by /qa-orchestrator.

Create custom QA agent personalities for project-specific testing needs. Guided builder that asks about the specialty, tools, and test scenarios, then generates a personality file and registers it in the QA config. Trigger on "create a QA personality", "add a custom test agent", "build a webhook tester", or when the user needs a project-specific QA agent. Also triggered by /qa-create-personality.

Generate, evaluate, audit, and normalize QA test cases to RAVN standards. Trigger on "generate/write/create test cases", "evaluate/score my test cases", "audit my test suite", "review test coverage", "normalize/reformat test cases", or when a user wants test design help. Also triggered by /testcases.

Generate professional QA Test Plan documents (.docx or .pdf) from a structured interview. Trigger on "create/write a test plan", "I need a test plan", "prepare QA documentation", /testplan, or when a user uploads a PRD/requirements and wants a test plan generated.

Framework-agnostic testing principles — test philosophy, structure, mocking boundaries. Use when writing, reviewing, or debugging tests.

Vitest testing framework for modern JavaScript. Unit and integration tests with Jest-compatible API, native TypeScript, Vite integration, MSW v2 for HTTP mocking, and snapshot testing. Covers vi.mock hoisting, fake timers, test isolation, and async patterns. Use when writing tests with Vitest, mocking dependencies, or setting up test infrastructure.

Universal code quality rules — KISS, DRY, clean code, code review. Base skill every project should include. Use when writing or reviewing any code.

TypeScript language patterns and type safety rules — strict mode, no any, discriminated unions. Use when writing TypeScript code, reviewing types, or enforcing type safety.

Set up and enforce a strict, production-grade ESLint configuration for TypeScript projects, then systematically fix all linting issues. Use this skill whenever the user asks to add a linter or ESLint, enforce code quality rules, fix linting errors, clean up code style, or add type-aware linting. Trigger on: "lint", "eslint", "code quality", "static analysis", "strict linting", "make it stricter", "make the code stricter", "add better rules", "clean up the codebase", "enforce standards", "fix all the warnings", or "ShadCN lint errors". Handles detection, config generation, dependency installation, auto-fix, and manual remediation. Do NOT use for Biome or Rome projects, Prettier-only formatting, non-TypeScript/JavaScript projects, writing custom ESLint rules or plugins, husky/lint-staged/pre-commit hook setup, or when the user just wants to run an existing linter without changing its configuration.