git clone https://github.com/benjam3n/reasoningtool.git
cd reasoningtool/claude-code-plugin
claude
cp -r path/to/reasoningtool/claude-code-plugin/skills your-project/
cp path/to/reasoningtool/claude-code-plugin/CLAUDE.md your-project/
cd your-project
claude
Codex
git clone https://github.com/benjam3n/reasoningtool.git
cd reasoningtool
./src/scripts/install_codex_skills.sh
Installs all skills into ~/.codex/skills using symlinks.
Use --copy to copy instead of symlink. Use --force to replace existing installed skills.
/araw should I change careers
/se what are all the ways I could approach this product launch
/aex we need to hire 3 more engineers by Q2
/cmp React vs Svelte for this project
/rca our deploy pipeline keeps failing
/uaua 4x what's the best approach to learning math
You have…
Use
What it does
A claim
/claim
ARAW tests both sides
A decision
/decide
Compares options via ARAW
An idea
/viability
ARAW viability test
Work to assess
/evaluate
Correctness + risk check
Something broken
/diagnose
UAUA cause mapping
A space to explore
/search
UAUA exploration
“How do I?”
/how
Method discovery
A goal
/want
Commitment tracing
A feeling
/emotion
Acknowledge + route
You have…
Use
What it does
Something to execute
/action
Plans and executes
Content to produce
/create
Routes to writing skills
Need max effort
/certainty
Iterates until resolved
Something to improve
/iterate
Meta-iteration
Domain question
/technical
Specialized routing
Analysis needed
/analyze
Decomposition + synthesis
Need orientation
/meta
Skill discovery
Improve a prompt
/sp
Adds precision + criteria
What do you have?
├── A statement → testable? → /claim or /viability
├── A question → /decide, /how, /diagnose, or /search
├── A goal → /want
├── A feeling → /emotion
├── Work to review → /evaluate
├── Something to do → /action
├── Content to write → /create
├── Technical question → /technical
├── Need analysis → /analyze
├── Not sure → /meta
Skill
Example
What you get
/araw
/araw claim
Default (2x) — 4–5 levels, 20+ findings
/araw 4x claim
5–6 levels, 35+ findings
/araw 8x claim
6–8 levels, 55+ findings
/uaua
/uaua 4x topic
Alternating search at 4x
/se
/se topic
Systematic exploration
/rca
/rca problem
Root cause analysis
/ht
/ht assumption
Hypothesis testing
pip install openai pyyaml
cd src/araw
python auto_expand_llm.py --db career.db --seed "Should I start a business?"
python visualize.py --db career.db --serve
pip install openai pyyaml
export OPENAI_API_KEY="sk-your-key"
cd src/araw
# Start a new exploration
python auto_expand_llm.py --db my_analysis.db --seed "Should I start a business?"
# Continue expanding
python auto_expand_llm.py --db my_analysis.db --continue --parallel 5
# View in browser
python visualize.py --db my_analysis.db --serve
# Open http://localhost:8080
python araw_engine.py create "I need to change careers" my_search.db
python araw_engine.py stats my_search.db
python araw_engine.py export my_search.db
python auto_expand_llm.py --db career.db --seed "I should quit my job"
python auto_expand_llm.py --db career.db --continue --parallel 5 --max-depth 6
python auto_expand.py --db quick.db --seed "I must learn to code" --duration 60