A prompt-writing skill for setting up long-running autonomous coding-agent sessions. It produces a detailed mission, work plan, checks, and finish conditions for an agent that will work without repeated questions.

Use it to prepare overnight or multi-hour coding runs, define what is in scope, assign work across agents, require testing and review, and specify a provable stopping point.

It turns a broad project idea into clear boundaries and verifiable completion criteria, reducing drift during unattended work.

loop-maker — Skill by jbrazy480

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 skills/jbrazy480/loop-maker/loop-maker

Any agent

npx skills add jbrazy480/loop-maker --skill loop-maker

Clone the repo

git clone --depth 1 https://github.com/jbrazy480/loop-maker

Made for: Claude Code, Codex.

Per session 177 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 3,561 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.

Model	Per session	Once invoked
Fable 5	$0.00177	$0.03561
Opus 5	$0.00088	$0.01781
Sonnet 5	$0.00035	$0.00712
Haiku 4.5	$0.00018	$0.00356

Measured 2d ago against content hash 0b5011f83812, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

loop-maker 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.

loop-maker/SKILL.md · 147 lines

How it starts

The opening of the file, as written. The whole thing — 147 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Loop Maker

You write loops, not prompts. When someone tells you what they want, you produce a complete, safe autonomous run they can launch and walk away from — the kind that ships real, verified work over 3-5+ hours without going off the rails and without ever stopping to ask a question.

Built on the loop-engineering pattern (Peter Steinberger, Boris Cherny, Geoff Huntley's Ralph loop, Anthropic's long-running-harness research) plus the plan-delegate-review orchestration method: the smartest model never builds anything — it plans, delegates to cheaper workers, and reviews. That's how a whole company gets built for ~500K orchestrator tokens.

The non-negotiable rule

A loop is only as safe as its weakest of three parts. Never ship a loop missing any one of them:

The Spec (the pin) — what to build, what's explicitly OUT of scope, acceptance criteria. Stops the agent from inventing scope.
The Verifier (not the writer) — the work is checked by something other than the agent that wrote it: tests, lint, typecheck, a screenshot review, a fresh-context skeptic agent. The writer never grades its own homework.
The Stop Condition (provable) — a finish line the transcript can literally prove, plus a turn/time cap as the safety net. Never "make it good" — that's unprovable, and the loop will either never end or lie that it's done.

If the request can't be given a provable stop condition yet, your FIRST job is to help define one.

Audience note

Many users arrive from an Instagram giveaway and may be new to agents or coding. Warm, plain English, zero unexplained jargon. If someone just types "loop," reply with one friendly line ("Tell me what you're building and I'll write you a loop that runs it for you — even 'a landing page for my coffee shop' works"), not a form. The magic moment is realizing they don't have to babysit the AI anymore.

Explain yourself as you go — in super simple words, every time

The skill teaches by narrating. At each step, add ONE plain-English line a total beginner would get — never a lecture, never jargon without a 5-word translation:

Read the full file on GitHub · 147 lines

Files

What ships with it

13 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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.

2d ago First seen · 147 lines · 177 tokens per session scan A 0b5011f83812

Subscribe to this mod's changes

loop-maker is a skill published in the GitHub repository jbrazy480/loop-maker (27 stars, last pushed 1mo ago), licensed MIT. It adds 177 tokens to every session and 3,561 once invoked, about $0.0009 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 2026-08-30.

Autonomous Agents Code review Frontend Testing agent-skills ai-agents automation claude claude-code

hive.chart-creation-foundations

Required reading whenever any chart tool is available. Teaches the one-tool embedding contract (call chartrender → live chart appears in chat AND a downloadable PNG lands in the queen session dir), the ECharts (data viz) vs Mermaid (structural diagrams) decision, the BI/financial-grade aesthetic baseline (no…

aden-hive/hive · 133 tokens

browser-edge-cases

SOP for debugging browser automation failures on complex websites. Use when browser tools fail on specific sites like LinkedIn, Twitter/X, SPAs, or sites with Shadow DOM.

aden-hive/hive · 40 tokens

canvas-design

Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.

netease-youdao/LobsterAI · 59 tokens

loop-maker