pr-review-fix-loop

An iterative command for reviewing a pull request and fixing selected code issues. A pull request is a proposed code change for review before it is merged into a project.

In plain words
What is it for?
Reviewing and fixing Ruby, Node, Python, Go, or Rust changes, with optional tests, linting, selected review aspects, iteration limits, and Codex-based review.
Why use it?
It repeats review and correction so important findings can be addressed and checked again, instead of stopping after one review.

Command

Part of the pr-review-fix-loop plugin — 1 skill, 2 commands, 2 hooks shipped together

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/dapi/claude-code-marketplace/pr-review-fix-loop
Clone the repo
git clone --depth 1 https://github.com/dapi/claude-code-marketplace

Or install pr-review-fix-loop, the plugin that ships this one along with the rest of its 1 skill, 2 commands, 2 hooks.

Per session 42 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,113 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 $0.00042 $0.04113
Opus 5 $0.00021 $0.02056
Sonnet 5 $0.00008 $0.00823
Haiku 4.5 $0.00004 $0.00411

Measured 3d ago against content hash bcce29f15957, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

pr-review-fix-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 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.

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.

pr-review-fix-loop/commands/pr-review-fix-loop.md · 294 lines

How it starts

The opening of the file, as written. The whole thing — 294 lines — stays where its author put it; the contents beside it link to each section on GitHub.

PR Review Fix Loop

Итеративный цикл: запустить PR review, исправить критические и важные замечания, повторить до чистого отчёта.

Парсинг аргументов

Разобрать $ARGUMENTS:

  • --max-iterations N — максимум итераций (по умолчанию 20)
  • --aspects ASPECTS — аспекты ревью (по умолчанию code errors tests)
  • --min-criticality N — минимальный уровень criticality для исправления, 1-10 (по умолчанию 5)
  • --lint — запускать линтер с автофиксом после исправлений (по умолчанию выключен). Линтер определяется автоматически по типу проекта
  • --codex — параллельно запускать ревью через Codex CLI (по умолчанию выключен)
  • --base BRANCH — base branch для Codex diff (по умолчанию: автодетект из PR или main branch из CLAUDE.md)

Доступные аспекты: code, errors, tests, comments, types, simplify, all

Валидация аргументов

  • --max-iterations должен быть целым числом >= 1. Если значение невалидно — сообщить пользователю ошибку и прекратить.
  • --min-criticality должен быть целым числом от 1 до 10 включительно. Если меньше 1 — установить 1, если больше 10 — установить 10, сообщить пользователю об автокоррекции.
  • --aspects — каждый указанный аспект должен входить в список доступных. Если найден неизвестный аспект — сообщить пользователю ошибку со списком доступных аспектов и прекратить. При парсинге убрать обрамляющие кавычки, если они есть.
  • Если --base указан без --codex — предупредить пользователю что --base используется только вместе с --codex и будет проигнорирован.

Проверки перед запуском

Проверка .gitignore

Перед всеми остальными проверками запустить "${CLAUDE_PLUGIN_ROOT}/scripts/check-gitignore.sh". Если action_needed равно true, добавить отсутствующие файлы в .gitignore немедленно. Продолжить с остальными проверками.

Версия и детект проекта

Получить версию и параметры проекта одним вызовом:

VERSION=$(jq -r '.version' "${CLAUDE_PLUGIN_ROOT}/.claude-plugin/plugin.json") && PROJECT_JSON=$("${CLAUDE_PLUGIN_ROOT}/scripts/detect-project.sh") && echo "pr-review-fix-loop v${VERSION}" && echo "$PROJECT_JSON"

Read the full file on GitHub · 294 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 · 294 lines · 42 tokens per session scan A bcce29f15957

Subscribe to this mod's changes

pr-review-fix-loop is a command published in the GitHub repository dapi/claude-code-marketplace (16 stars, last pushed 4mo ago), licensed MIT. It adds 42 tokens to every session and 4,113 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 2026-08-30.