Simon Zimmermann
Community@SimonZimmer · Düsseldorf, Germany
Simon Zimmermann builds token-efficient coding skills for Claude Code and OpenCode, covering compressed communication, code review, commits, and repository mapping.
Agent Skills by Simon Zimmermann
Showing 14 vetted skills indexed across 1 GitHub repositories.
SimonZimmercaveman-help
Displays a quick-reference card of caveman modes, skills, and configuration options.
SimonZimmercaveman-review
Generates one-line code review comments with location, severity, problem, and fix.
SimonZimmeroh-my-opencode-slim
Configure agents, models, prompts, skills, and MCPs for oh-my-opencode-slim in OpenCode.
SimonZimmercaveman-compress
Compress natural language memory files into caveman format to reduce input tokens.
SimonZimmerreflect
Analyzes recent OpenCode sessions to recommend reusable skills, agents, commands, and config changes.
SimonZimmercodemap
Generate hierarchical codemaps and change-aware state files for unfamiliar repositories.
SimonZimmerclonedeps
Clone dependency source repositories into an ignored local workspace for agent inspection.
SimonZimmerworktrees
Orchestrates Git worktrees as isolated coding lanes for parallel and risky development work.
SimonZimmerverification-planning
Plans evidence paths for verifying non-trivial code changes before implementation.
SimonZimmercaveman-stats
Reports actual session token usage and estimated savings from the Claude Code session log.
SimonZimmerdeepwork
Orchestrates multi-phase coding workflows with delegated specialists and mandatory review gates.
SimonZimmersimplify
Simplifies code for readability while preserving exact behavior and project conventions.
SimonZimmercaveman
Compresses AI responses into terse caveman-style prose while preserving technical accuracy.
SimonZimmercaveman-commit
Generates terse Conventional Commits messages from staged changes.
Frequently Asked Questions About Simon Zimmermann
FAQPage SchemaWhat tasks can I accomplish with Simon Zimmermann's skills?▼
You can compress prompts and memory files to save tokens, generate ultra-compact code review comments and commit messages, map unfamiliar codebases hierarchically, clone dependency sources for inspection, manage Git worktrees, and plan verification for non-trivial coding changes.
Who are these skills designed for?▼
They target developers using Claude Code or OpenCode who want lower token costs and tighter output. Engineers doing frequent PR reviews, commits, large refactors, or onboarding onto unfamiliar repositories benefit most from the compression, codemap, and verification-planning skills.
How do I trigger and use these skills in practice?▼
Skills activate via slash commands like /caveman, /caveman-compress FILEPATH, /caveman-review, and /caveman-stats, or natural phrases such as 'review this PR' or 'compress memory file'. Several auto-trigger, for example caveman-review on pull requests and caveman-commit when staging changes.
How does the caveman mode reduce token usage?▼
Caveman mode speaks in ultra-compressed language, cutting token usage by roughly 75% while keeping full technical accuracy. It supports intensity levels lite, full, ultra, and wenyan variants. The caveman-stats skill reads the Claude Code session log to show real usage and savings.
What prerequisites or dependencies do these skills require?▼
Skills assume a Claude Code or OpenCode environment. caveman-stats reads the Claude Code session log directly, clonedeps requires a local workspace for dependency sources, worktrees needs a Git repository, and caveman-compress overwrites files while saving a FILE.original.md backup.