Commands Reference

Complete reference for all Mannay slash commands.

Overview

Commands are quick tools for scaffolding and common tasks. Use /command-name syntax.

API Commands

/api-new

Create a new API endpoint with validation and error handling.

/api-new POST /api/users create user endpoint

Creates:

• API route file
• TypeScript types
• Validation schema
• Error handling
• Test file

/api-backend-setup

Set up backend API with Express or Fastify for Vite/CRA projects.

/api-backend-setup express
/api-backend-setup fastify

Creates:

• Server configuration
• Middleware setup
• Route structure
• Error handling

/api-test

Generate comprehensive test suite for an API endpoint.

/api-test /api/users

Creates:

• Happy path tests
• Error case tests
• Edge case tests
• Integration tests

/api-protect

Add authentication and authorization to API endpoints.

/api-protect /api/admin

Adds:

• Auth middleware
• Role-based access
• Token validation
• Security headers

UI Commands

/component-new

Create a new React component with TypeScript.

/component-new UserCard

Creates:

• Component file
• TypeScript props interface
• Test file
• Storybook story (optional)

/page-new

Create a new Next.js page with App Router.

/page-new dashboard

Creates:

• Page component
• Layout (if needed)
• Loading state
• Error boundary

/page-new-react

Create a React Router page for Vite/CRA projects.

/page-new-react settings

Creates:

• Page component
• Route registration
• TypeScript types

Supabase Commands

/types-gen

Generate TypeScript types from Supabase database schema.

/types-gen

Creates:

• Database types
• Table interfaces
• Function types

/edge-function-new

Create a new Supabase Edge Function.

/edge-function-new send-email

Creates:

• Edge function file
• Deno configuration
• Type definitions

Quality Commands

/lint

Run ESLint, TypeScript, and Prettier.

/lint

Runs:

• ESLint check
• TypeScript check
• Prettier check
• Auto-fix when possible

/code-cleanup

Refactor and clean up code following best practices.

/code-cleanup src/utils/helpers.ts

Does:

• Remove dead code
• Improve naming
• Apply DRY
• Add missing types

/code-optimize

Analyze and optimize code for performance.

/code-optimize src/components/List.tsx

Does:

• Profile performance
• Identify bottlenecks
• Apply optimizations
• Measure improvement

/code-explain

Explain complex code sections.

/code-explain src/lib/auth.ts

Provides:

• Line-by-line explanation
• Architecture overview
• Key patterns used

Planning Commands

/new-task

Analyze task complexity and create implementation plan.

/new-task add user authentication

Creates:

• Task breakdown
• Complexity estimate
• Implementation plan
• Agent recommendations

/feature-plan

Create technical specification for a feature.

/feature-plan user profile management

Creates:

• User stories
• Technical spec
• API design
• Implementation phases

/docs-generate

Generate documentation for code, APIs, or components.

/docs-generate src/lib/

Creates:

• API documentation
• Code comments
• README files

Ralph Commands

/ralph-init

Initialize Ralph Mode files in current project.

/ralph-init

Creates:

• scripts/ralph/ directory
• prd.json template
• progress.txt
• prompt.md
• AGENTS.md

/ralph-build

Create a PRD interactively for Ralph execution.

/ralph-build

Process:

1. Ask clarifying questions
2. Generate user stories
3. Create prd.json
4. Validate atomicity

/ralph-start

Start Ralph autonomous execution loop.

/ralph-start

Does:

1. Reads prd.json
2. Executes stories sequentially
3. Commits after each story
4. Updates progress

/ralph-status

Check Ralph execution status.

/ralph-status

Shows:

• Stories completed
• Stories remaining
• Current progress
• Recent commits

/ralph-stop

Stop Ralph execution loop.

/ralph-stop

Does:

• Terminates loop
• Preserves progress
• Reports status

Codex Commands

/gpt-review

Queue a file for GPT code review using your ChatGPT subscription via Codex CLI.

/gpt-review src/auth.ts --security
/gpt-review src/api/payment.ts --performance
/gpt-review src/utils.ts --refactor
/gpt-review src/file.ts "Check for edge cases"

Flags: | Flag | Focus | |——|——-| | --security | Vulnerabilities, input validation, secrets | | --performance | Bottlenecks, memory leaks, optimization | | --refactor | DRY, naming, structure, clarity |

Requirements:

• Codex CLI installed: npm install -g @openai/codex
• Authenticated: codex auth (Sign in with ChatGPT)

/gpt-status

Check Codex review status and daily usage.

/gpt-status

Shows:

• Reviews completed today
• Daily limit remaining
• Recent review list

/gpt-results

View GPT code review results.

/gpt-results              # Show all recent results
/gpt-results src/auth.ts  # Show specific file's review

Shows:

• Summary of each review
• Full results saved in .claude/gpt-reviews/

Memory Commands

/memory-status

Check memory system status and statistics.

/memory-status

Shows:

• Database size
• Saved handoffs
• Cached signatures
• Token savings
• Learnings count
• Current session state

/memory-learnings

View recent learnings accumulated across sessions.

/memory-learnings

Shows:

• Recent learnings
• Tags and categories
• Recall counts
• Total learnings

See Memory System for full documentation.

Command Quick Reference