A Claude skill that gives Claude deep knowledge of DSPy.rb for building type-safe LLM applications in Ruby using Sorbet types, composable modules, and ReAct/CodeAct agents.
git clone https://github.com/vicentereig/dspy-rb-skill.gitThe DSPy.rb Claude Skill equips Claude with deep knowledge of the DSPy.rb framework, enabling it to assist Ruby developers in building type-safe LLM applications. Once installed, Claude can help define signatures with Sorbet types including enums, structs, and arrays, and guide developers in composing modular LLM workflows. The skill covers creating ReAct and CodeAct agents with tool calling, implementing recursive types using JSON Schema $defs format, and adding field descriptions to T::Struct for improved LLM understanding. It also supports optimization workflows using MIPROv2 and GEPA, as well as testing with RSpec and VCR and deploying to production with observability and error handling. Ruby developers building AI-powered applications with DSPy.rb are the primary beneficiaries of this skill.
For Claude Code, clone the repository into your global skills directory at ~/.claude/skills/dspy-rb or into a project-level .claude/skills/dspy-rb folder. For Claude.ai Pro, Max, Team, or Enterprise plans, download the repository as a ZIP and upload it via Settings > Skills. Once installed, the skill activates automatically when you ask about DSPy.rb, Ruby LLM applications, signatures, predictors, or ReAct agents.
Define type-safe DSPy signatures with Sorbet enums, structs, and arrays
Create ReAct and CodeAct agents with tool calling in Ruby
Implement LLM pipeline optimization using MIPROv2 and GEPA
Test and evaluate LLM applications with RSpec and VCR
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/vicentereig/dspy-rb-skillCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
I'm developing a Ruby application using DSPy.rb for type-safe LLM integration. I need help with [SPECIFIC TASK] in my project. Here's my current code: [CODE SNIPPET]. Can you help me [IMPROVE/DEBUG/EXTEND] this functionality while maintaining type safety?
```ruby
# Type-safe LLM integration example for a customer support chatbot
class SupportChatbot < DSPy::LLM
# Define input and output types
input_type :string, description: "User's support query"
output_type :string, description: "Generated support response"
# Initialize with specific LLM configuration
def initialize(llm_config: { model: "gpt-3.5-turbo", temperature: 0.7 })
super(llm_config)
end
# Custom processing method
def process_query(query)
# Type-safe validation
raise ArgumentError, "Query must be a string" unless query.is_a?(String)
# Generate response with type safety
response = generate(
prompt: "Answer this support query: #{query}",
max_tokens: 150
)
# Post-processing with type checking
if response.is_a?(String)
"Support Response: #{response}"
else
raise "Invalid response type received"
end
end
end
# Usage example
chatbot = SupportChatbot.new
response = chatbot.process_query("How do I reset my password?")
puts response
```Real-time collaborative writing platform
AI assistant built for thoughtful, nuanced conversation
neobank for e-commerce businesses
Cloud ETL platform for non-technical data integration
Customer feedback management made simple
Enterprise workflow automation and service management platform
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan