A set of rules for preventing an AI agent from repeating failed tool calls or getting stuck in a loop. It requires the agent to inspect errors, use new information, change approach, and stop after repeated failures.

Guiding tool use during file searches, edits, and other operations where repeated errors could cause an endless cycle.

It reduces wasted time when file edits, searches, or other tool actions fail. The agent is prompted to learn from each failure instead of retrying the same action unchanged.

antiloop — Skill for Claude Code and Codex by Major-Woolfi

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.

Any agent

npx skills add Major-Woolfi/Skills_for_AI-Agents --skill antiloop

Clone the repo

git clone --depth 1 https://github.com/Major-Woolfi/Skills_for_AI-Agents

Made for: Claude Code, Codex.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/major-woolfi/skills_for_ai-agents/antiloop/github.svg)](https://agentmods.dev/skills/major-woolfi/skills_for_ai-agents/antiloop)

Your own site

<a href="https://agentmods.dev/skills/major-woolfi/skills_for_ai-agents/antiloop"><img src="https://agentmods.dev/badge/skills/major-woolfi/skills_for_ai-agents/antiloop/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/major-woolfi/skills_for_ai-agents/antiloop"><img src="https://agentmods.dev/badge/skills/major-woolfi/skills_for_ai-agents/antiloop.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 49 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 948 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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 5.1	$0.00049	$0.00948
Opus 5.5	$0.00020	$0.00379
Sonnet 5.5	$0.00010	$0.00190
Haiku 4.5	$0.00005	$0.00095

Measured 2mo ago against content hash f8350ceab1f1, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

antiloop 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 2mo 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.

SKILLS/antiloop/SKILL.md · 109 lines

How it starts

The opening of the file, as written. The whole thing — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AntiLoop — Навык предотвращения зацикливаний

Описание

Навык для предотвращения зацикливаний при использовании инструментов и размышлении. Применяется автоматически при любой работе с файлами и инструментами записи.

Когда активировать

Автоматически при любом использовании edit_file, run_terminal_command
При любой ошибке поиска или изменения
При повторных неудачных попытках вызова инструментов

Философия

Один промах — не ошибка. Два промаха — повод остановиться. Три — сигнал к смене стратегии.

Принципы AntiLoop

Ограничение на повторения. Один и тот же вызов — не более двух раз подряд.
Перечитай перед повтором. Каждый повтор обязан использовать новые данные.
Меняй стратегию. Если два подхода не сработали — пробуй третий.
Останавливайся. Три шага без прогресса — сигнал к остановке.

Обязательный цикл

Шаг 1: Первая попытка

Вызови инструмент с наилучшими аргументами.

Шаг 2: Анализ ошибки

Если инструмент вернул ошибку — не повторяй тот же вызов.

Перечитай relevant-часть файла через read_file
Найди точное содержимое для old_string / path / query
Определи, почему предыдущая попытка не сработала

Шаг 3: Корректировка

Вызови инструмент с скорректированными аргументами.

Шаг 4: Смена стратегии (если нужно)

Если инструмент не сработал второй раз:

edit_file не нашёл old_string → перечитай файл, найди точное совпадение, проверь отступы
grep_search не нашёл паттерн → попробуй другой паттерн, используй file_glob_search
read_file вернул меньше строк → смести line_start, читай дальше
run_terminal_command вернул ошибку → проверь путь, права, синтаксис

Шаг 5: Остановка

Если 3 шага подряд не дали прогресса:

Прекрати циклические вызовы
Верни пользователю краткий статус
Предложи альтернативу или задай уточняющий вопрос

Чек-лист

Я перечитал файл и знаю точное содержимое
Мои аргументы отличаются от предыдущих
Я понимаю, почему предыдущая попытка не сработала
Я знаю, что буду делать при следующей неудаче

Read the full file on GitHub · 109 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.

2mo ago First seen · 109 lines · 49 tokens per session scan A f8350ceab1f1

Subscribe to this mod's changes

antiloop is a skill published in the GitHub repository Major-Woolfi/Skills_for_AI-Agents (2 stars, last pushed 1mo ago), licensed MIT. It adds 49 tokens to every session and 948 once invoked, about $0.0002 per session on Opus 5.5. 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.

Building agents

antiloop next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
antiloopMajor-Woolfi/Skills_for_AI-Agents	Skill	2	1mo ago	MIT	A	49	Claude Code, Codex
imagegenopenai/codex	Skill	133k	yesterday	Apache-2.0	A	113	Codex
playwright-climicrosoft/playwright	Skill	99k	yesterday	Apache-2.0	A	19	Claude Code
local-ai-agentsmicrosoft/ai-agents-for-beginners	Skill	79k	4d ago	MIT	A	200	Claude Code, Codex
systematic-debuggingobra/superpowers	Skill	313k	3d ago	MIT	A	21	Claude Code
next-cache-components-adoptionvercel/next.js	Skill	143k	yesterday	MIT	A	95	Claude Code
next-partial-prefetching-adoptionvercel/next.js	Skill	143k	yesterday	MIT	A	103	Claude Code