Synthesizer Docs — Development skill for Claude Code
You are an expert at writing Claude Code CLAUDE.md files.
How to install Synthesizer Docs
Installs to ~/.claude/skills/joelbarmettleruzh-ultrainit.sh-synthesizer-docs/SKILL.md
mkdir -p ~/.claude/skills/joelbarmettleruzh-ultrainit.sh-synthesizer-docs && curl -fsSL https://raw.githubusercontent.com/joelbarmettlerUZH/ultrainit.sh/HEAD/prompts/synthesizer-docs.md -o ~/.claude/skills/joelbarmettleruzh-ultrainit.sh-synthesizer-docs/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Synthesizer Docs does
You are an expert at writing Claude Code CLAUDE.md files. You receive exhaustive codebase analysis and must produce comprehensive, deeply-structured CLAUDE.md files.
This is Pass 1 of 2. You produce ONLY the documentation artifacts: root CLAUDE.md and subdirectory CLAUDE.md files. Skills, hooks, and agents are handled in Pass 2 by another agent.
Core Principles
- Evidence over opinion. Every line must trace to the findings.
- Dense, not brief. Be as long as needed — a 300-line CL
Alternatives in Development
- Scientific Agent Skills — A set of ready to use Agent Skills for research, science, engineering, analysis, finance and writing 18.7k ★
- Use Google Gamini — Using Gemini CLI for Large Codebase Analysis 220 ★
- Coder Eval Create Plan — Create a structured, phased implementation plan for a feature or change in the coder_eval codebase, executable 116 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Synthesizer Tooling
You are an expert at creating Claude Code skills, hooks, and subagents. You receive exhaustive codebase analys
Synthesizer
You are an expert at creating Claude Code configurations. You receive exhaustive codebase analysis from multip
Claude Code Pua
A Claude Code skill that puts the coding agent on a performance improvement plan: structured failure escalatio
Workflows Review
Perform exhaustive code reviews using multi-agent analysis, ultra-thinking, and worktrees
Clavain Review
Perform exhaustive code reviews using multi-agent analysis and deep inspection
Ce Review
Perform exhaustive code reviews using multi-agent analysis, ultra-thinking, and worktrees
Related Agents
Codebase Mapper
Use to produce structured analysis documents of a codebase — stack, architecture, conventions, testing, or con
Gsd Assumptions Analyzer
Deeply analyzes codebase for a phase and returns structured assumptions with evidence. Spawned by discuss-phas
Thrunt Scope Analyzer
Deeply analyzes codebase for a phase and returns structured assumptions with evidence. Spawned by shape-hypoth