skill-learning-patterns
Skill Learning Patterns
This meta-skill teaches agents to recognize valuable learnings during their work and contribute improvements back to the communal skill repository through pull requests.
Core Philosophy
This repository is a living knowledge base that improves through collective learning. As agents work on tasks, they discover:
- Better approaches than documented
- Missing error handling cases
- Gaps in existing skills
- Clearer ways to explain patterns
- New patterns worth capturing
These discoveries should flow back into the repository so all agents benefit.
Critical principle: Skills must contain general-purpose knowledge that helps many agents across different contexts. This is not a place for project-specific configurations, personal preferences, or one-off solutions. Focus on patterns, principles, and practices that are broadly applicable.
When to Use This Skill
More from oleksandrkucherenko/e-bash
pester
PowerShell TDD testing framework guidance for Pester v5+. Use when writing, structuring, or debugging PowerShell unit tests; mocking cmdlets, native commands (bash, git, curl), or .NET types; isolating tests with TestDrive/TestRegistry; capturing output streams; generating code coverage or JUnit/NUnit reports for CI/CD; running parameterized or tagged tests; or troubleshooting Pester Discovery vs Run phase issues.
19elegant-code
Language-agnostic rulebook for producing, reviewing, and improving elegant code. Use when writing new code, refactoring existing code, reviewing code quality, or when user asks for "elegant", "clean", "maintainable", or "well-structured" code. Applies to any programming language or framework.
14gemini-cli
Use Gemini CLI as a complementary AI tool for tasks requiring massive context windows (1M tokens). Invoke when analyzing large codebases, requesting deep analysis with extended thinking, getting second opinions on complex problems, or when Claude's context limits are insufficient. Triggers include phrases like "use gemini", "analyze with gemini", "get second opinion", "deep analysis of codebase", or when processing files exceeding Claude's context capacity.
1