prompt-optimization

rems-buddy

Rem's buddy — reusable Claude Code skills packaged as a plugin marketplace. Install once, use across all projects.

Installation

Add to your Claude Code plugins:

/plugins add /Users/remihuguet/workspaces/rems-buddy

Or add to .claude/settings.json:

{
  "plugins": ["/Users/remihuguet/workspaces/rems-buddy"]
}

Available Plugins

bugfix — TDD Bug Fix (command)

Strict red-green workflow: write a failing test reproducing the bug, implement minimal fix, run full suite, commit.

/bugfix users can log in with expired tokens

git-workflows — Conventional Commits (command + skill)

Commands: /commit, /commit-push

Always-on skill: conventional commit message formatting rules.

coding-standards — Naming and Comments (skill)

Always-on guidance for domain naming, intent-revealing identifiers, comment strategies, and minimal docstrings.

python-testing — Testing Conventions (skill)

Three skills providing always-on guidance:

• pytest-conventions — Function-style tests, fixtures, parametrize, exception testing
• testing-philosophy — Behavior-driven testing, diamond approach, fakes over mocks
• testing-organization — Three-tier strategy: unit, integration, e2e

python-architecture — Backend Architecture Patterns (skill)

Six skills for Python backend architecture:

• hexagonal-layers — Five-layer architecture (domain, service, adapters, views, entrypoints)
• domain-driven-design — Value objects, entities, aggregates, repositories, events, commands
• cqrs — Command/query separation
• event-driven — MessageBus, UnitOfWork, handler patterns
• architecture-pitfalls — Common anti-patterns to avoid
• file-organization — Module organization principles

prompt-optimization — Intent Clarity (skill)

Always-on prompt interpretation: resolve ambiguity from context, decompose vague requests, minimize unnecessary questions.

Plugin Structure

plugin-name/
├── .claude-plugin/
│   └── plugin.json          # Plugin manifest (name, description, version, author)
├── commands/                 # On-demand slash commands
│   └── my-command.md
├── skills/                   # Always-on behavioral guidance
│   └── my-skill/
│       └── SKILL.md
└── README.md

Adding New Plugins

1. Create a plugin directory following the structure above
2. Add .claude-plugin/plugin.json with name, description, version, and author
3. Place skills under skills/<skill-name>/SKILL.md
4. Place commands under commands/<command-name>.md with YAML frontmatter
5. Register the plugin in .claude-plugin/marketplace.json
6. Skill paths in marketplace.json are relative to the plugin source (e.g., ./skills/my-skill)