12og3r/redpen/setup

Claude Code and agent skill by 12og3r/redpen

Trust tier 2
  • community
LLM review pending
Source check pending

Checking that the SKILL.md file is still in the repository.

Configure redpen for the ChatGPT desktop launcher and Codex (language + native-style hint). Invoke when the user wants to change which language they are practising or whether the native-style hint shows. Writes ~/.codex/redpen.config.

Install

npx vskill@latest install 12og3r/redpen/setup

SKILL.md

The user invoked the redpen-setup skill. Follow these steps EXACTLY. Do not explore the codebase, do not run other commands, do not summarise.

Note on model: the runner automatically tries gpt-5.6-luna in Fast
mode, then gpt-5.6-terra in Fast mode, and finally gpt-5.4-mini in
Standard mode. The setup skill therefore does NOT ask about model.

Step 1 — Read current config

Read ~/.codex/redpen.config. Parse:

  • LANGUAGE=<...> (default: english)
  • SHOW_HINT=<on|off> (default: on)

Remember as CURRENT_LANGUAGE and CURRENT_SHOW_HINT.

Missing-config case: if the file does not exist yet, treat both as unset (no ✓ markers in Step 2). The defaults above only apply if the user picks an out-of-range answer in Step 3.

Step 2 — Ask the user (two questions, one at a time)

Ask Question 1 and wait for the user's reply. Only after they reply, ask Question 2. Do NOT batch the two questions into one turn — the user needs to see and answer each one independently.

Current-selection marker: append ✓ to the option whose value matches the user's current config (so they see what they're on). The ✓ is independent of (Recommended); both can appear together. When matching the answer in Step 3, accept EITHER the number OR the label, and ignore any trailing ✓.

Question 1 — language

Reply with the number or the name. Which language do you want the redpen plugin to coach you on?

  1. English
  2. 中文 (Chinese)
  3. Español (Spanish)
  4. 日本語 (Japanese)

Question 2 — native style line

Reply with the number or the name. Show a second "native style" line with a more idiomatic rephrasing under each rewrite?

  1. On (Recommended) — show the divider + native-style rephrasing.
  2. Off — only show the scored rewrite line.

Step 3 — Map answers to config values

Language (accept number OR name, case-insensitive, ignore trailing ✓):

  • 1 or English → english
  • 2 or 中文 / Chinese / 中文 (Chinese) → chinese
  • 3 or Español / Spanish / Español (Spanish) → spanish
  • 4 or 日本語 / Japanese / 日本語 (Japanese) → japanese
  • Anything else → english

Native hint (accept number OR name, case-insensitive):

  • 1 or On → on
  • 2 or Off → off
  • Anything else → on

Step 4 — Write the new config

Use the Write tool to overwrite ~/.codex/redpen.config with EXACTLY this content (substitute the chosen values):

# redpen config — shared by the ChatGPT desktop launcher and Codex hook.
#
# Supported LANGUAGE values:
# english | chinese | spanish | japanese
# aliases: en | zh, cn, 中文 | es, español, espanol | ja, jp, 日本語
LANGUAGE=<new language>
#
# SHOW_HINT: whether to show a second "native style" line under each rewrite
# with a more idiomatic, colloquial rephrasing. on (default) | off.
SHOW_HINT=<new show_hint>
#
# NOTE: The runner automatically falls back through:
# gpt-5.6-luna Fast -> gpt-5.6-terra Fast -> gpt-5.4-mini Standard.

Step 5 — Confirm

Reply with ONE short line summarising what changed. Examples:

  • Both changed: ✓ Switched to 中文 with native-style hints off.
  • Only language: ✓ Language set to English (hints unchanged: on).
  • Only hint: ✓ Native-style hints off (language unchanged: English).
  • Nothing changed: Already on English + hints on — no change.

New setting takes effect on your next prompt. No further explanation.

Security

  • Security scanPassed, 100 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
56
Repository forks
4
7-day trend score
0

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/12og3r/redpen/setup/badge)](https://verified-skill.com/skills/12og3r/redpen/setup)

Something wrong with this skill or its page? Tell us.