← 返回 Skills 市场

Huo15 Openclaw Plan Mode

Name: Huo15 Openclaw Plan Mode
Author: zhaobod1

作者 Job Zhao · GitHub ↗ · v1.0.3 · MIT-0

cross-platform ✓ 安全检测通过

197

总下载

当前安装

版本数

在 OpenClaw 中安装

/install huo15-openclaw-plan-mode

功能描述

结构化规划模式 — 在执行复杂任务前先做系统性规划。借鉴 Claude Code 的 Plan Agent。

使用说明 (SKILL.md)

规划模式 (Plan Mode)

在执行复杂或多步骤任务前，进入结构化规划模式。

使用时机

✅ 使用此技能当：

用户请求涉及多个步骤的复杂任务
"帮我规划一下..."、"设计一个方案..."
需要在动手前理清思路
修改涉及多个文件或系统

❌ 不要使用当：

简单的单步操作（改个名字、修个 typo）
用户明确要求直接执行

规划流程

阶段一：理解需求

复述需求 — 用自己的话总结用户要做什么
识别约束 — 有什么限制条件？时间、兼容性、依赖？
提出澄清问题 — 如果有不确定的地方，先问清楚再规划

阶段二：调研现状

读相关代码 — 理解现有实现，不要凭空设计
找可复用的 — 搜索已有的函数、工具、模式
识别风险 — 哪些地方可能出问题？

阶段三：设计方案

输出结构化方案：

## 背景
为什么要做这个改动？解决什么问题？

## 方案
### 步骤 1: [描述]
- 修改文件: path/to/file
- 具体操作: ...
- 复用已有: function_name from file

### 步骤 2: [描述]
...

## 风险与回退
- 风险1: ...  → 缓解措施: ...

## 验证方式
- [ ] 怎么确认改动正确？
- [ ] 运行什么测试？

阶段四：确认执行

将方案展示给用户
等待用户确认后再开始执行
执行过程中逐步标记完成

核心原则

先理解，再设计，最后执行 — 不要一上来就写代码
方案要具体 — 具体到哪个文件哪个函数，不要泛泛而谈
复用优先 — 能用已有的就不要重新造
最小变更 — 只改需要改的，不顺手重构

安全使用建议

This skill is coherent and small-surface: it creates structured plans and asks the agent to read and reference code files. Before installing or running it, consider: (1) keep it user-invocable (don't grant blanket automatic execution), (2) ensure the agent's file access is scoped to the intended project/workspace so it can't read unrelated sensitive files, (3) always review the generated plan and changes before authorizing automated edits, and (4) if you want stronger guarantees, run the skill in a sandboxed environment or require explicit confirmation from you before any write operations are performed.

功能分析

Type: OpenClaw Skill Name: huo15-openclaw-plan-mode Version: 1.0.3 The skill bundle 'huo15-openclaw-plan-mode' consists solely of instructional markdown (SKILL.md) designed to guide an AI agent through a structured planning workflow. It encourages best practices such as requirement clarification, code research, and obtaining user confirmation before executing complex tasks, with no executable code, network calls, or suspicious instructions present.

能力评估

✓ Purpose & Capability

The name/description (structured planning for complex tasks, inspired by a 'Plan Agent') matches the SKILL.md content: it defines a multi-phase planning workflow and explicit expectations for producing file-level, step-by-step plans. No unrelated binaries, env vars, or installs are requested.

ℹ Instruction Scope

The instructions explicitly tell the agent to 'read related code', identify files/functions, and produce/describe concrete file edits. That behavior is coherent for a planning agent that prepares code changes, but it implies the agent may access project files and propose modifications — users should ensure the agent only has access to the intended project workspace and that the plan is reviewed before any automated edits are applied.

✓ Install Mechanism

No install spec or code files are present (instruction-only). This minimizes disk-write and supply-chain risk.

✓ Credentials

The skill requires no environment variables, credentials, or config paths. The SKILL.md does not reference secret env vars or external services, so requested access is proportionate to its purpose.

✓ Persistence & Privilege

always is false and the skill is user-invocable. It does not request persistent presence or system-wide configuration changes. Autonomous invocation is allowed by default but not a red flag on its own here.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install huo15-openclaw-plan-mode
安装完成后，直接呼叫该 Skill 的名称或使用 /huo15-openclaw-plan-mode 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.3

- No file or content changes detected in this version. - Version number updated to 1.0.3.

v1.0.2

- Updated version number in SKILL.md from 1.0.0 to 1.0.2. - No other content or functionality changes made.

v1.0.1

- Clarified structured planning process for complex, multi-step tasks before execution. - Added detailed phase breakdown: understanding requirements, researching status quo, designing solutions, and confirming execution with the user. - Included clear guidelines for when to use or avoid plan mode. - Provided step-by-step plan output format with background, actions, risk mitigation, and validation. - Emphasized core principles: understand first, design specifically, maximize reuse, and minimize changes.

v1.0.0

- 首发版本，提供结构化规划模式，用于多步骤或复杂任务前的系统性规划 - 借鉴 Claude Code 的 Plan Agent，包括需求分析、现状调研、方案设计及执行确认流程 - 明确规划适用与非适用场景，强调先规划后执行，确保改动有据可依 - 输出结构化的规划模板，包含背景、具体步骤、风险与回退、验证方式等内容 - 核心原则：优先理解和复用、最小变更、明确步骤

元数据

Slug huo15-openclaw-plan-mode

版本 1.0.3

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 4

常见问题