Borrowing it
Nothing to install: this file belongs to akiojin/llmlb. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/akiojin/llmlb/develop/.claude/commands/ralph-loop.mdgit clone --depth 1 https://github.com/akiojin/llmlbWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/akiojin/llmlb/ralph-loop)<a href="https://agentmods.dev/commands/akiojin/llmlb/ralph-loop"><img src="https://agentmods.dev/badge/commands/akiojin/llmlb/ralph-loop/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/akiojin/llmlb/ralph-loop"><img src="https://agentmods.dev/badge/commands/akiojin/llmlb/ralph-loop.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.00352 |
| Opus 5.5 | $0.00000 | $0.00141 |
| Sonnet 5.5 | $0.00000 | $0.00070 |
| Haiku 4.5 | $0.00000 | $0.00035 |
Grade A, and why
ralph-loop 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 8d 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.
What it actually says
Ralph Wiggum Loop
Start Ralph Wiggum loop in current session (project)
User Input
$ARGUMENTS
Instructions
This command starts an autonomous execution loop using the "Ralph Wiggum technique" - a method for continuous task execution without requiring user confirmation between steps.
Execution Flow
- Parse the user's task: Understand what needs to be accomplished from the user input above
- Create a comprehensive todo list: Break down the task into actionable items using TodoWrite
- Execute tasks autonomously: Work through each task without asking for confirmation
- Self-continue: After completing each task, immediately proceed to the next one
- Report completion: Only stop when all tasks are complete or an unrecoverable error occurs
Rules
- No confirmation required: Proceed with implementation without asking "Should I continue?"
- Self-driving: After each step, immediately start the next one
- Error handling: If a task fails, attempt to fix it. Only stop if truly stuck.
- Progress tracking: Keep the todo list updated as you work
- Quality checks: Run necessary checks (tests, linting) as part of the workflow
- Commit when appropriate: Create commits at logical checkpoints
Starting the Loop
Begin by:
- Reading any relevant spec files (if a SPEC ID was provided)
- Understanding the current project state
- Creating a detailed todo list
- Starting execution immediately
GO NOW - Start working on the task described in the user input. Do not ask questions. Execute autonomously until complete.
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.
- 8d ago First seen · 42 lines · 0 tokens per session scan A 4ef7bd4a93f2
ralph-loop is a command published in the GitHub repository akiojin/llmlb (10 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 352 tokens. 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 2026-10-21.
Other commands, from other repositories
ws
Word summary — anything told at N levels, from N words down to 1 word, for anything: a topic ("russian literature"), a URL, YouTube link, file, pasted text, a question, or "this". A leading number sets N (5 by default). Shortcut for the Fabric skill's WordSummary intent.
cancel-ralph
Cancel active Ralph Loop.
engage.actions
Execute Phase 7 - Actions on Objectives and Goal Achievement.
r
Resume the last handoff and carry on with the outstanding work.
focus
You are an ADHD focus assistant. Help the user commit to ONE task and block mental distractions.
stuck
You are an ADHD-friendly unblocking assistant. Help users get unstuck when they don't know what to do.