A golang development helper with AI-powered features and customizable prompt templates. Don't memorize commands when you can mcq lint
mcq help
$ mcq help
This application provides shortcuts to common development tasks
Usage:
mcq [command]
Available Commands:
ai AI-powered commands
all Run everything
build -> go build
ci Run almost everything
clean -> fmt deps vet
cover -> go tool cover
deps -> go mod tidy, download, vendor
docker docker build, run, push
fmt -> go fmt
help Help about any command
install -> go install
lint -> golangci-lint, staticcheck
log -> ~git log --graph --oneline --decorate --all
run -> go run main.go
setup install dependencies
test -> go test
version Version
Flags:
-h, --help help for mcq
Use "mcq [command] --help" for more information about a command.
Convert vague feature requests into detailed user stories using AI models and copy to clipboard.
Setup:
# For Claude (Anthropic)
export ANTHROPIC_API_KEY='your-anthropic-api-key-here'
# For GPT-4o (OpenAI)
export OPENAI_API_KEY='your-openai-api-key-here'
# Or set both to choose between modelsUsage:
# Auto-detect model based on available API keys
mcq ai jira "Add a dark mode to the application"
# Specify model explicitly
mcq ai jira --model claude "Add a dark mode to the application"
mcq ai jira --model gpt-4o "Add a dark mode to the application"
mcq ai jira --model gpt-5 "Add a dark mode to the application"
mcq ai jira --model gpt-5-mini "Add a dark mode to the application"
mcq ai jira --model gpt-5-nano "Add a dark mode to the application"
# Set verbosity levels for structured logging
mcq ai jira --verbosity 1 "Add a dark mode to the application" # Basic logging
mcq ai jira --verbosity 2 "Add a dark mode to the application" # Detailed logging
mcq ai jira --verbosity 3 "Add a dark mode to the application" # Verbose logging (includes streaming details)Supported Models:
- Claude Sonnet 4.5 (Anthropic) - Latest Claude model
- GPT-4o (OpenAI) - Previous generation GPT model
- GPT-5 (OpenAI) - Latest and most advanced GPT model
- GPT-5 Mini (OpenAI) - Faster and more cost-effective GPT-5 variant
- GPT-5 Nano (OpenAI) - Optimized for simple tasks with minimal latency
This will generate a user story in the format "As a [user type], I want [goal] so that [benefit]" with additional acceptance criteria and technical considerations, then copy it to your clipboard.
For more details, see AI_FEATURE.md.
Customize AI prompts using Go templates without modifying the code.
Setup:
# Generate example template files
mcq templates generate ./my-templates
# Set custom template directory
export MCQ_PROMPTS_DIR='./my-templates'
# Validate templates
mcq templates validateAvailable Commands:
mcq templates generate [dir]- Generate example template filesmcq templates validate- Validate template syntaxmcq templates list- List available prompt types
Template Variables:
{{.FeatureRequest}}- User's feature request{{.RepositoryContext}}- Repository information{{.ProjectName}},{{.ModulePath}},{{.GoVersion}}- Project details{{.Readme}},{{.RecentCommits}},{{.Dependencies}}- Context data{{.Now}}- Current timestamp
For more details, see TEMPLATES.md.
- Mechanism to fail fast during commands running. If error, it should quit. (OrderedRunner)
- Mechanism for pretty printing text to screen. Likely a writer library/passed around with global defaults for different types of messages
- Mechanism for parallelization of tasks than can be completed together
- Simplify colorwriter