onboarding

An onboarding guide for RACA, a research-assistant system for running experiments on computing clusters. It helps set up clusters, Hugging Face, and the dashboard.

In plain words
What is it for?
It welcomes users, connects computing clusters, configures Hugging Face, and launches the research dashboard.
Why use it?
It gives new users a starting path through the required setup instead of leaving them to discover each configuration step themselves.

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/onboarding
Clone the repo
git clone --depth 1 https://github.com/Zayne-sprague/Dr-Claude-Code

Made for: Claude Code.

Per session 18 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,183 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 2 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 $0.00018 $0.02183
Opus 5 $0.00009 $0.01092
Sonnet 5 $0.00004 $0.00437
Haiku 4.5 $0.00002 $0.00218

Measured 3d ago against content hash fc8eea24c97b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

onboarding scanned grade B with 2 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 3d 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.

Unrestricted tool accessmediumExcessive agency

A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.

**YOUR FIRST MESSAGE MUST BE THE WELCOME GREETING. Do not run any commands, read any files, or do any background work before greeting the user.** Just say hello.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -s -o /dev/null -w "%{http_code}" http://127.0.0.1:7860/
.claude/commands/raca/onboarding.md · 225 lines

How it starts

The opening of the file, as written. The whole thing — 225 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.

YOUR FIRST MESSAGE MUST BE THE WELCOME GREETING. Do not run any commands, read any files, or do any background work before greeting the user. Just say hello.

State Tracking

State lives at .raca/onboarding_state.json. Read it only AFTER you've greeted the user (not before). Update after every step.

{
  "step": "welcome",
  "clusters": [],
  "hf_configured": "pending",
  "dashboard_local": "pending",
  "completed": false,
  "dashboard_url": null,
  "hf_org": null,
  "updated_at": null
}

NEVER mention the state file, steps, or tracking to the user.

Resume

If onboarding was already started (state file exists and step != "welcome"), summarize briefly:

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


Step 1: Welcome

Greet the user immediately. No tool calls before this message.

"Hey, welcome to RACA! I'm your research assistant — I help you design experiments, run jobs on clusters, and keep everything organized."

"This folder is your new research home. All your experiments, code, notes, and results live here:"

  • private_projects/ — your research code (training scripts, eval pipelines)
  • notes/ — experiment tracking + personal notes (I read these for context)
  • packages/ — shared code across experiments
  • tools/ — the dashboard, CLI, and other tooling

"If you hit any errors during setup, just copy paste them here and I'll help you through it!"

"Let's start by connecting your compute. Do you have access to any of these?"

  1. A SLURM cluster (university HPC, national lab, etc.)
  2. RunPod (cloud GPUs on demand)
  3. I don't have remote compute yet — that's fine, we can add clusters later.

Step 2: Compute Setup

If they have clusters, set them up. If not, skip to Step 3.

"How many clusters do you want to connect?"

For each cluster:

  1. Ask: SLURM cluster or RunPod?

Read the full file on GitHub · 225 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. 3d ago First seen · 225 lines · 18 tokens per session scan B fc8eea24c97b

Subscribe to this mod's changes

onboarding is a command published in the GitHub repository Zayne-sprague/Dr-Claude-Code (5 stars, last pushed 5mo ago), licensed MIT. It adds 18 tokens to every session and 2,183 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 2 findings (unrestricted tool access, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.