跳转到主要内容
此内容尚未提供您的语言版本,正在以英文显示。

Ralph

技能 已验证 活跃

Self-referential loop until task completion with configurable verification reviewer

目的

To guarantee complex tasks are fully completed and verified according to defined user stories, preventing silent failures and partial implementations.

功能

  • PRD-driven task execution
  • Iterative story-by-story completion
  • Automatic retry on failure
  • Mandatory reviewer verification
  • Session persistence and state tracking

使用场景

  • Ensuring guaranteed task completion with verification.
  • Managing complex tasks that may span multiple iterations.
  • Tasks requiring structured progress tracking and reviewer sign-off.
  • When user explicitly requests 'don't stop' or 'finish this'.

非目标

  • Full autonomous pipeline from idea to code (use Autopilot).
  • Exploration or planning before committing (use Plan skill).
  • Quick one-shot fixes (delegate directly to an executor agent).
  • Manual control over completion (use Ultrawork directly).

Documentation

  • info:Configuration & parameter referenceKey parameters like `--no-deslop` and `--critic` are mentioned, but a comprehensive reference for all options and their defaults is not explicitly detailed.

Code Execution

  • info:ValidationWhile the skill relies on structured prompts and agent delegation, explicit mention of schema-based validation for all inputs and outputs is not detailed.

安装

请先添加 Marketplace

/plugin marketplace add Yeachan-Heo/oh-my-claudecode
/plugin install oh-my-claudecode@omc

质量评分

已验证
95 /100
1 day ago 分析

信任信号

最近提交1 day ago
星标33.7k
许可证MIT
状态
查看源代码