Token导航 LogoToken导航TokenDH.com
研究检索只读github未标认证来源可访问许可证需确认审计通过

deep-plan深度计划

Agent Skill

deep-plan 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

727

周安装

30

GitHub Stars

14

下载量

238
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:deep-plan(深度计划)
来源仓库:https://github.com/theclaymethod/deep-plan
仓库路径:skills/deep-plan
安装命令:
npx skills add https://github.com/theclaymethod/deep-plan --skill deep-plan
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/theclaymethod/deep-plan --skill deep-plan

简介

deep-plan 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中快速定位候选结果。

  • 适用于需要根据关键词或任务场景进行信息筛选的场景,如项目规划支持。
  • 通过关键词输入和来源线索,Agent 可生成结构化候选列表供进一步核验。
  • 安装命令为 npx skills add https://github.com/theclaymethod/deep-plan --skill deep-plan。
  • 使用前需确认权限范围和维护状态,注意是否触发联网或文件操作。

SKILL.md

Deep Plan

Disciplined workflow that prevents wasted effort by separating thinking from typing. Never write code until the user has reviewed and approved a written plan.

Workflow

Research → Plan → [Interview] → Annotate (1-6x) → Todo List → Implement → Feedback → Archive

[Interview] is optional — an interactive Q&A to resolve ambiguities before annotation. See Phase 2 reference for triggers and process.

Critical guard: Do NOT advance to the next phase unless the user explicitly says to. When in doubt, ask.

Phases

Each phase has detailed instructions in its reference file. Read the relevant file when entering that phase.

PhaseFileArtifactGuard
1. Researchreferences/phase-1-research.md.claude/research.mdWait for user review
2. Planreferences/phase-2-plan.md.claude/plan.mdDo not implement yet
3. Annotatereferences/phase-3-annotate.mdUser edits plan.mdDo not implement yet
4. Todo Listreferences/phase-4-todo.mdCheckboxes in plan.mdWait for user approval
5. Implementreferences/phase-5-implement.mdMark tasks [x]Run typecheck continuously
6. Feedbackreferences/phase-6-feedback.mdTerse correctionsRevert if directionally wrong
7. Archivereferences/phase-7-archive.mddocs/completed-tasks/

Templates

Use these when creating artifacts:

  • Research: assets/templates/research-template.md
  • Plan: assets/templates/plan-template.md

Scripts

ScriptPurposeWhen to Use
scripts/check_annotations.py <file>Find unaddressed user annotationsBefore updating plan in Phase 3
scripts/plan_progress.py <file>Report task completion (X/Y)During Phase 5 implementation
scripts/archive_plan.py <file> --name <slug>Move plan to docs/completed-tasks/Phase 7 archive

Reference Files

FilePurpose
references/annotation-guide.mdHow users write effective annotations
references/phase-*.mdDetailed instructions per phase

Hard Constraints

These are non-negotiable. Violating any of these is a workflow failure.

ConstraintApplies To
Never write code without an approved plan. No exceptions.All phases
"Don't implement yet" is absolute. Do not start coding during research, planning, or annotation phases, even if the plan looks complete.Phases 1-4
All research goes into a persistent file. Never summarize findings only in chat. Write .claude/research.md.Phase 1
Read source files before proposing changes. Every code snippet in the plan must reference real files you've actually read. Never design in a vacuum.Phase 2
Address every annotation. Do not skip, deprioritize, or silently drop any user note.Phase 3
Do not use any or unknown types. Maintain strict typing throughout.Phase 5
Do not add unnecessary comments or jsdocs. Code should be self-documenting.Phase 5
Run typecheck continuously. Catch problems during implementation, not after.Phase 5
Do not stop until all tasks are completed. Don't pause mid-implementation for confirmation unless blocked.Phase 5
Mark tasks complete in the plan as you go. The plan document is the single source of truth for progress.Phase 5
Scope is what the plan says. If the user cut something, it stays cut. Don't sneak it back in. Don't add unrequested features.All phases

Key Principles

  • The plan is shared mutable state. Both you and the user edit it. It persists across compaction.
  • Single long sessions. Research through implementation in one conversation.
  • Implementation should be boring. Creative decisions happen in annotation cycles.
  • Never speculate about unread code. If the plan references a file, read it first.
  • Scope discipline. If the user cuts something from the plan, it's cut. Don't sneak it back in.

Credit

This workflow is based on How I Use Claude Code by Boris Tane.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

35.58%
按下载量换算85

Claude

26.52%
按下载量换算63

Cursor

17.88%
按下载量换算43

Gemini CLI

10.14%
按下载量换算24

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills