← 返回 Skills 市场
Executing Plans
作者
chenleiyanquan
· GitHub ↗
· v0.1.0
7351
总下载
7
收藏
132
当前安装
1
版本数
在 OpenClaw 中安装
/install executing-plans
功能描述
Use when you have a written implementation plan to execute in a separate session with review checkpoints
安全使用建议
Safe to install as a lightweight workflow skill. Use it only with implementation plans you trust, read each checkpoint before approving more work, and review the referenced finishing-a-development-branch skill separately because final behavior depends on it.
功能分析
Type: OpenClaw Skill
Name: executing-plans
Version: 0.1.0
The skill bundle is benign. The `_meta.json` file contains standard metadata. The `SKILL.md` file provides instructions for an AI agent to execute a development plan in batches, emphasizing critical review, verification, reporting, and stopping when blocked. There are no instructions for data exfiltration, malicious execution, persistence, or prompt injection aiming for harmful objectives. The use of a sub-skill (`superpowers:finishing-a-development-branch`) is a standard modular approach and not indicative of malicious intent.
能力评估
Purpose & Capability
The skill says it executes a written implementation plan in batches with review checkpoints, and the artifact does exactly that: load the plan, review it, execute a small batch, verify, report, and wait for feedback.
Instruction Scope
It tells the agent to follow plan steps exactly, which makes the plan itself important to trust, but it also requires critical review, verification, checkpoint reporting, and stopping when blocked or unclear.
Install Mechanism
The package contains only SKILL.md and no executable installer or scripts. It does require a disclosed finishing-a-development-branch sub-skill at the end, so that referenced skill should be reviewed separately.
Credentials
No credentials, environment variables, network access, package installs, broad local indexing, or sensitive data access are requested by this artifact.
Persistence & Privilege
The artifact shows no background workers, persistence mechanism, privilege escalation, exfiltration path, destructive automation, or hidden behavior.
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 executing-plans - 安装完成后，直接呼叫该 Skill 的名称或使用
/executing-plans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
v0.1.0
- Initial release of the executing-plans skill.
- Introduces a structured process: load and review plan, execute tasks in batches, and report progress for feedback.
- Emphasizes critical review, stopping for clarification on blockers, and not guessing when instructions are unclear.
- Includes explicit instructions for batch size, checkpoints, and required skill handoff for finishing development.
- Clearly defines when to pause for help or revisit previous steps based on plan changes or feedback.
元数据
常见问题
Executing Plans 是什么？
Use when you have a written implementation plan to execute in a separate session with review checkpoints. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 7351 次。
如何安装 Executing Plans？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install executing-plans」即可一键安装，无需额外配置。
Executing Plans 是免费的吗？
是的，Executing Plans 完全免费（开源免费），可自由下载、安装和使用。
Executing Plans 支持哪些平台？
Executing Plans 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Executing Plans？
由 chenleiyanquan（@chenleiyanquan）开发并维护，当前版本 v0.1.0。
推荐 Skills