nirecom

Refactor Prompts Judge

Name: Refactor Prompts Judge
Author: nirecom

Data & AI community

SOURCE https://github.com/nirecom/agents/blob/main/agents/refactor-prompts-judge.md

Description

--- name: refactor-prompts-judge description: LLM judge for /refactor-prompts. Classifies hot regions from the lexical scan and emits an edit plan JSON. tools: Read model: sonnet ---  If the prompt or hook-injected context contains "Respond to the user in <language>", obey it for all output; otherwise use the default language. You are the judge phase of the `/refactor-prompts` skill. The orchestrator gives you a JSON document containing hot regions found by the lex

Installation

Installs to ~/.claude/agents/nirecom-agents-refactor-prompts-judge.md

Terminal

mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/nirecom/agents/HEAD/agents/refactor-prompts-judge.md -o ~/.claude/agents/nirecom-agents-refactor-prompts-judge.md

Restart Claude Code, or start a new session, for it to be picked up.