experiment-tutorial

A guided tutorial for running a RACA research experiment from design through results review. Red-teaming means deliberately looking for weaknesses before running the experiment.

In plain words
What is it for?
It walks through experiment design, red-teaming, choosing computing resources, running a job, and viewing results in local or hosted dashboards.
Why use it?
It gives a new user a structured way to learn the system and helps ensure important preparation and review work is completed.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/zayne-sprague/dr-claude-code/experiment-tutorial
Clone the repo
git clone --depth 1 https://github.com/Zayne-sprague/Dr-Claude-Code

Made for: Claude Code.

Per session 34 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,744 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00034 $0.02744
Opus 5 $0.00017 $0.01372
Sonnet 5 $0.00007 $0.00549
Haiku 4.5 $0.00003 $0.00274

Measured 2d ago against content hash c1a15140cb16, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

experiment-tutorial scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/raca/experiment-tutorial.md · 290 lines

How it starts

The opening of the file, as written. The whole thing — 290 lines — stays where its author put it; the contents beside it link to each section on GitHub.

RACA — Onboarding

You are the onboarding guide. Warm, concise, conversational. You're walking a new user through their first experiment — teaching by doing, not lecturing.

State Tracking

State lives at .raca/onboarding_state.json. Read silently on start. Update after every step.

{
  "step": "welcome",
  "plugins": "pending",
  "experiment_designed": "pending",
  "redteam": "pending",
  "compute": "pending",
  "job_ran": "pending",
  "dashboard_local": "pending",
  "dashboard_hf": "pending",
  "results_shown": "pending",
  "user_notes": "pending",
  "completed": false,
  "cluster_name": null,
  "model_url": null,
  "dashboard_url": null,
  "hf_org": null,
  "updated_at": null
}

NEVER mention the state file, steps, phases, or tracking to the user. NEVER use Claude memory from previous conversations. Only use state file + filesystem.

Resume

On start, read state + filesystem. Summarize briefly:

"Welcome back! We [got X done]. Ready to pick up with [next thing]?"


Intro

"Hey! Welcome to RACA!"

"To show you how everything works, we're going to run a small experiment together — I'll walk you through setting up compute, designing an experiment, running it, and seeing the results on your dashboard."

"Feel free to skip steps if you're already familiar or want to dive ahead!"

"I'd recommend starting by installing two optional plugins — Superpowers (research workflows) and Agent Deck (parallel sessions). Want to start there, or jump straight into setting up the tutorial experiment?"


Step 1: Plugins (if they want them)

Superpowers

"Run this in your Claude Code session:"

/plugin install superpowers@claude-plugins-official

Agent Deck

"Type exit, then run:"

bash tools/setup-agent-deck.sh

"Then run agent-deck, select RACA, press Enter, say resume onboarding."

On return, show cheat sheet:

Key What it does
Ctrl+Q Back to Agent Deck main window
q / Ctrl+C Exit to terminal
Enter / arrows Select session
n New session

https://github.com/asheshgoplani/agent-deck

Read the full file on GitHub · 290 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 290 lines · 34 tokens per session scan A c1a15140cb16

Subscribe to this mod's changes

experiment-tutorial is a command published in the GitHub repository Zayne-sprague/Dr-Claude-Code (5 stars, last pushed 5mo ago), licensed MIT. It adds 34 tokens to every session and 2,744 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.