Fluent Korean 🤖 Agent Open source
Claude Code output style that makes the agent write clear, natural Korean
- GitHub stars
- 1.4k
- Stars this week
- +26
- Forks
- 95
- Licence
- MIT
- Last push
- 2026-08-23
- Maintainer
- snflkd
/plugin marketplace add snflkd/fluent-koreanThird-party subagents & agents run with your permissions. Read the source before installing, and prefer pinned versions.
Works with
About Fluent Korean
What it does
Fluent Korean is a small Claude Code plugin that ships output styles for writing clear Korean. Coding agents are often tuned to be terse, and in Korean that tends to produce dropped particles and verb endings, strings of bare nouns and odd vocabulary that are hard to read and easy to misread. This plugin adds instructions at the system-prompt level that tell the model to restore those elements and prefer plain, commonly used words, aiming for unambiguous meaning over elegant prose. The author, who studied Korean literature, wrote the guideline text and the Korean parts of the README by hand.
What is inside
fluent-korean: an output style that keeps Claude Code's coding instructions, for normal development workfluent-korean-not-coding: a variant with the coding instructions removed, for sessions where Claude is not editing code- Optional add-on lines in the README for extra behaviour, such as explaining things for beginner developers, using honorifics, avoiding rare dictionary words, applying the rules to all Korean output, or always answering in Korean
- Guidance for using the same text as custom instructions in Claude's web and desktop apps or in other AI tools
Works with
It is packaged as a Claude Code plugin. Because the core is a Markdown instruction file, it can be pasted into CLAUDE.md, project instructions or other assistants' settings, though that is a manual step.
How to install
/plugin marketplace add snflkd/fluent-korean
/plugin install fluent-korean@fluent-korean
After installing, pick one of the two styles in /config and start a new session or run /clear.
Maintenance and safety
The plugin is MIT-licensed, recently updated and contains only instruction text, with no scripts. The author notes that it uses slightly more tokens, that adherence weakens in long or instruction-heavy sessions, and that subagent behaviour should be checked. The README is mostly in Korean with a short English summary.
Who should use it
Korean-speaking developers who prompt Claude Code in Korean or produce Korean documentation, commit messages or reports and find the default output hard to read. It does nothing for English-only work.
Pros
- Coding and non-coding variants of the output style
- Instruction text can be reused in other tools
- No scripts, just Markdown instructions
Cons
- Only useful for Korean-language work; README mostly in Korean
- Adds some tokens and adherence fades in long sessions
Similar subagents & agents
All agent workflows & frameworks →Spec Kit 🤖 AgentFree
GitHub's toolkit for spec-driven development with AI coding agents
Task Master 🤖 AgentFree
AI task management that turns a PRD into tasks your coding agent works through
Fast Agent 🤖 AgentOpen source
Python framework for building, orchestrating and evaluating MCP-native AI agents
OpenSpec 🤖 AgentFree
Lightweight spec-driven development: agree on changes before the agent codes
CC Safety Net 🤖 AgentOpen source
Pre-execution guard that blocks destructive commands and secret access for coding agents
BMAD Method 🤖 AgentFree
Agile AI-driven development with analyst, PM, architect, developer and UX agents