Skills Plugins MCP Prompt Model 博客 我的中心

grill-me

Conducts a relentless interview about every aspect of a plan or idea, walking down each branch of the design tree until shared understanding is reached. Invoke with /grill-me before starting any significant design or implementation work.

DeepseekModel 官方收录技能 质量 良好 · 48 v1.0.0

获取

https://deepseekmodel.com/api/download.php?id=tokugero-opencode-project-template-template-claude-skills-grill-me-skill-md&format=skill
下载 .skill 标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用
.skill 文件中 system_prompt 字段的实际内容。
name grill-me description Conducts a relentless interview about every aspect of a plan or idea, walking down each branch of the design tree until shared understanding is reached. Invoke with /grill-me before starting any significant design or implementation work. context fork allowed-tools ["Read","Glob","Grep"] grill-me Drives an exhaustive design interview before implementation begins. Surfaces ambiguities, forces decisions on every branch of the design tree, and resolves dependencies between decisions one at a time — so that implementation work starts from a shared understanding rather than unchecked assumptions. When to use Invoke with /grill-me before starting any significant feature, architecture decision, or refactor. Also useful when a plan exists but feels underspecified. Inputs Plan or idea (required): Describe what you want to build or decide. Can be vague — the interview will sharpen it. Steps Ask the user to describe the plan or idea if not already provided. Interview relentlessly about every aspect of the plan until a shared understanding is reached: "Interview me relentlessly about every aspect of this plan until we reach a shared understanding. Walk down each branch of the design tree, resolving dependencies between decisions one by one. And finally, if a question can be answered by exploring the codebase, explore the codebase instead." Work through the design tree systematically: Identify all major decision branches (data model, interfaces, error handling, performance, security, rollback, testing strategy, etc.) For each branch, ask one focused question at a time Resolve each answer before moving to the next branch When a question is answerable from the codebase, read the relevant files (L0 → L1 → L2) instead of asking After all branches are resolved, produce a concise Design Summary : The decision made on each branch Any assumptions that remain (explicitly flagged) Recommended next step (e.g., /write-a-prd , implementation task) Expected output A shared understanding of the full design, with all major decision branches resolved and a written Design Summary the user can reference during implementation.
Agent 识别该技能的关键词,点击任意一个即可复制。

该技能未提供触发词。

下载的 .skill 包内含以下字段。
字段 说明
format格式标识(skill/v1)
skill_id技能唯一 ID
name技能名称
version版本号
description技能描述
category所属分类(数组)
trigger_words触发词列表
tags标签列表
source来源标识
source_url来源链接(本页地址)
exported_at导出时间(每次下载生成)
system_prompt系统提示词正文
model_config模型参数:provider / model / temperature / max_tokens / top_p
examples示例
install_guide各平台导入说明(Coze / Dify / Claude / 自定义框架)
同一份技能可按不同平台格式导出。
.skill 标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用 下载
.skillpro 增强格式,额外含脚本 / 工具 / 依赖 / 钩子占位 下载
.json 纯 JSON 导出,只含 system_prompt 与模型参数 下载
Coze 带 frontmatter 的 Markdown,Coze 平台导入用 下载
Dify Dify DSL,创建应用后直接导入 下载

每日精选 Skill 推荐,免费送到你邮箱

输入邮箱,每天接收一个精选 AI Agent 技能推荐。完全免费,持续更新。

提交后我们会发送一封确认邮件,点击邮件里的链接才会开始收信。

完全免费,取消任意时间。我们不会发送垃圾邮件。