Skip to content

A language-agnostic collection of AI agent skills that enforce fundamental programming principles across **Cursor**, **Antigravity**, **GitHub Copilot**, and other AI coding assistants.

License

Notifications You must be signed in to change notification settings

Ariel-Rodriguez/programming-skills

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Programming Skills

Language-agnostic AI agent skills that enforce fundamental programming principles. This repository provides specific, granular instructions that enable AI coding assistants to produce significantly higher-quality code that adheres to robust engineering standards.

Adopting these skills measurably changes the output of AI models, shifting them from generating merely functional code to producing architecturally sound solutions.

Table of Contents

Installation

Select your platform for specific setup instructions:

How it Works

The core of this repository is the skills/ directory. Each skill is encapsulated in its own subdirectory following the ps-<name> convention (e.g., ps-composition-over-coordination).

We use this granular structure because:

  1. Focus: It allows the AI to load only the relevant context for a specific task, avoiding context window pollution.
  2. Modularity: Skills can be improved, versioned, and tested independently.
  3. Composability: Users can select the specific combination of principles they want to enforce for their project.

Validation & Testing

Every skill is validated against a rigorous testing suite found in the tests/ directory.

  • Automated Judging: We use an LLM-as-a-Judge approach. The system compares the output of a "Baseline" model (without the skill) against a "Skill" model (with the skill loaded).
  • Semantics over Syntax: The test does not just look for passing unit tests; it analyzes the logic and structure of the code.
  • Evidence-Based: The judge identifies the specific lines of code that demonstrate adherence to or violation of the principle.

Read our Case Study on Judge Fairness to see how the system fairly evaluates architectural quality, even when it means failing the Skill model.

Evaluation Results

Processed 24 evaluation(s).

Test Name Model Baseline With Skill Cases Pass Winner
results-ollama-devstral-small-2--24b-cloud-ps-composition-over-coordination devstral-small-2:24b-cloud good good ✅ 2/2 N/A
results-ollama-devstral-small-2--24b-cloud-ps-error-handling-design devstral-small-2:24b-cloud regular outstanding ✅ 2/2 With Skill
results-ollama-devstral-small-2--24b-cloud-ps-explicit-boundaries-adapters devstral-small-2:24b-cloud good outstanding ✅ 2/2 With Skill
results-ollama-devstral-small-2--24b-cloud-ps-explicit-ownership-lifecycle devstral-small-2:24b-cloud good good ✅ 2/2 With Skill
results-ollama-devstral-small-2--24b-cloud-ps-explicit-state-invariants devstral-small-2:24b-cloud good outstanding ✅ 2/2 With Skill
results-ollama-devstral-small-2--24b-cloud-ps-functional-core-imperative-shell devstral-small-2:24b-cloud regular good ✅ 2/2 With Skill
results-ollama-devstral-small-2--24b-cloud-ps-illegal-states-unrepresentable devstral-small-2:24b-cloud good outstanding ✅ 2/2 With Skill
results-ollama-devstral-small-2--24b-cloud-ps-local-reasoning devstral-small-2:24b-cloud good outstanding ✅ 2/2 With Skill
results-ollama-devstral-small-2--24b-cloud-ps-minimize-mutation devstral-small-2:24b-cloud good good ✅ 2/2 N/A
results-ollama-devstral-small-2--24b-cloud-ps-naming-as-design devstral-small-2:24b-cloud regular good ✅ 2/2 With Skill
results-ollama-devstral-small-2--24b-cloud-ps-policy-mechanism-separation devstral-small-2:24b-cloud good outstanding ✅ 2/2 With Skill
results-ollama-devstral-small-2--24b-cloud-ps-single-direction-data-flow devstral-small-2:24b-cloud regular good ✅ 2/2 With Skill
results-ollama-rnj-1--8b-cloud-ps-composition-over-coordination rnj-1:8b-cloud outstanding good ❌ 2/2 Baseline
results-ollama-rnj-1--8b-cloud-ps-error-handling-design rnj-1:8b-cloud vague outstanding ✅ 2/2 With Skill
results-ollama-rnj-1--8b-cloud-ps-explicit-boundaries-adapters rnj-1:8b-cloud regular outstanding ✅ 2/2 With Skill
results-ollama-rnj-1--8b-cloud-ps-explicit-ownership-lifecycle rnj-1:8b-cloud good outstanding ✅ 2/2 With Skill
results-ollama-rnj-1--8b-cloud-ps-explicit-state-invariants rnj-1:8b-cloud regular outstanding ✅ 2/2 With Skill
results-ollama-rnj-1--8b-cloud-ps-functional-core-imperative-shell rnj-1:8b-cloud regular outstanding ✅ 2/2 With Skill
results-ollama-rnj-1--8b-cloud-ps-illegal-states-unrepresentable rnj-1:8b-cloud outstanding outstanding ✅ 2/2 N/A
results-ollama-rnj-1--8b-cloud-ps-local-reasoning rnj-1:8b-cloud vague outstanding ✅ 2/2 With Skill
results-ollama-rnj-1--8b-cloud-ps-minimize-mutation rnj-1:8b-cloud regular outstanding ✅ 2/2 With Skill
results-ollama-rnj-1--8b-cloud-ps-naming-as-design rnj-1:8b-cloud vague good ✅ 2/2 With Skill
results-ollama-rnj-1--8b-cloud-ps-policy-mechanism-separation rnj-1:8b-cloud regular outstanding ✅ 2/2 With Skill
results-ollama-rnj-1--8b-cloud-ps-single-direction-data-flow rnj-1:8b-cloud vague good ✅ 2/2 With Skill

Documentation

License

MIT License - see LICENSE

About

A language-agnostic collection of AI agent skills that enforce fundamental programming principles across **Cursor**, **Antigravity**, **GitHub Copilot**, and other AI coding assistants.

Resources

License

Contributing

Stars

Watchers

Forks

Packages

No packages published