Token导航 LogoToken导航TokenDH.com
效率只读clawhub未标认证来源可访问clear审计通过

optional-strict-instructions可选的严格指示

Agent Skill

用于辅助提示词、系统指令、Agent 行为约束和工作流模板的整理。它适合让 Agent 规范任务边界、统一输出格式、拆分操作步骤或优化提示词可复用性。使用时需要保留真实业务约束,不要把示例当硬规则;涉及自动执行、外部工具或高风险操作时,应在提示词中明确确认步骤、权限边界和失败处理方式。

总安装

6,327

周安装

269

GitHub Stars

1

下载量

2,217
OpenClaw

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

MIT-0

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:optional-strict-instructions(可选的严格指示)
来源仓库:https://github.com/zhangwud1/optional-strict-instructions
安装命令:
openclaw skills install optional-strict-instructions
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 OpenClaw 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。

ClawHubOpenClaw
openclaw skills install optional-strict-instructions

简介

在执行敏感系统任务之前,通过强制用户确认、权限检查、明确的方法遵守和明确的选择来管理操作。

SKILL.md

name
optional-strict-instructions
description
Handle operations requiring user confirmation, permission verification, and strict adherence to explicit instructions. Activate when: (1) Operations need sudo/elevated permissions, (2) File deletion or modification is requested, (3) System changes are required, (4) User explicitly specifies a method (e.g., "use sudo"), (5) Sensitive operations need explicit consent, (6) Multiple valid approaches exist and user choice is required. Use this skill to ensure proper verification, user interaction, and strict compliance with user instructions.

Optional Strict Instructions Skill

This skill encapsulates the learned workflow for handling operations that require user confirmation, permission verification, and strict adherence to explicit instructions.

Core Principles Learned

1. Verify Before Acting

  • Always check resource existence first
  • Never assume user intent or permissions
  • Gather complete information before presenting options

2. Provide Clear Choices

  • Present multiple valid approaches
  • Include safety options (cancel, info-only, reversible actions)
  • Explain consequences and risks for each option
  • Structure options from most to least privileged

3. Wait for Explicit Confirmation

  • Never proceed without user input
  • Accept only clear, unambiguous choices
  • If input is unclear, ask for clarification

4. Follow Instructions Strictly

  • When user specifies a method (e.g., "use sudo"), use exactly that method
  • Do not substitute alternative approaches without permission
  • If specified method fails, report and wait for new instructions

5. Handle Authentication Properly

  • When sudo/authentication is required and fails, stop and report
  • Do not attempt unauthorized alternatives
  • Wait for password or alternative instruction

Standard Workflow

Phase 1: Verification

1. Check if target resource exists
2. Gather details (size, permissions, ownership, timestamps)
3. Determine what operations are possible

Phase 2: Option Presentation

[Operation Context]
Found: /path/to/resource (details)

Options:
1. [Method A - e.g., sudo with explanation]
2. [Method B - e.g., user permissions]
3. [Safety option - e.g., move to trash]
4. [Cancel operation]

Enter number choice (1-4):

Phase 3: Execution

1. Execute exactly as described in chosen option
2. If fails due to permissions/authentication, report and stop
3. Do not attempt unauthorized alternatives

Phase 4: Verification & Reporting

1. Verify operation completed successfully
2. Report results clearly
3. If failed, explain why and offer next steps

Key Scenarios & Patterns

Scenario 1: User Specifies Exact Method

User: "Use sudo to delete file.txt"

Correct Response:

  1. Check file.txt exists
  2. Execute: sudo rm file.txt
  3. If sudo needs password: "sudo requires password. Please provide password or choose alternative"
  4. Do NOT try rm file.txt without sudo

Wrong Response: Trying user permissions when sudo fails

Scenario 2: Multiple Valid Approaches

User: "Delete the log file"

Correct Response:

Found: /var/log/app.log (requires root)

Options:
1. sudo rm /var/log/app.log
2. sudo truncate /var/log/app.log (clear contents)
3. Show file contents first
4. Cancel

Scenario 3: Safety-Required Operations

User: "Modify system configuration"

Correct Response:

Modifying /etc/config/file (system-critical)

Options:
1. Proceed with backup first
2. Proceed without backup (risky)
3. Show proposed changes only
4. Cancel

Common Patterns

File Deletion Pattern

Found: /path/file (size, owner, modified)

Options:
1. sudo rm -f (permanent, needs password)
2. rm (permanent, user permissions)
3. trash/move to recycle (reversible)
4. Cancel

Sudo Operation Pattern

Operation requires sudo to [action]

Options:
1. Run with sudo (needs password)
2. Show command for manual execution
3. Check if alternative exists
4. Cancel

System Change Pattern

This will [describe change] affecting [system component]

Options:
1. Proceed with precautions [list]
2. Proceed without precautions (risk: [list])
3. Dry-run/show changes only
4. Cancel

Error Handling Rules

Rule 1: Authentication Failure

If sudo/authentication fails:
1. Report "Authentication required/failed"
2. Stop execution
3. Offer: "Provide password or choose alternative"
4. Do NOT attempt unauthorized methods

Rule 2: Permission Denied

If permission denied:
1. Report exact error
2. Explain why permission was denied
3. Offer appropriate alternatives
4. Do NOT attempt to bypass permissions

Rule 3: Resource Not Found

If resource doesn't exist:
1. Report "Resource not found: /path"
2. Suggest possible locations or alternatives
3. Do NOT proceed with operation

Learning from Mistakes

Mistake: Assuming Alternatives

Wrong: When sudo fails, try user permissions without asking Correct: Report failure, wait for instruction

Mistake: Over-automation

Wrong: Automatically choose "best" method Correct: Present options, let user choose

Mistake: Insufficient Verification

Wrong: Proceed without checking resource details Correct: Gather complete info first

Implementation Checklist

Before any sensitive operation:

  • [ ] Check resource exists
  • [ ] Gather permissions/ownership
  • [ ] Determine possible methods
  • [ ] Present clear options
  • [ ] Wait for user choice
  • [ ] Execute exactly as chosen
  • [ ] Verify results
  • [ ] Report completion

Reference Files

Remember

User choice > Automation efficiency Explicit instructions > Assumed intent Safety > Speed Verification > Assumption

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

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

补充不同宿主或平台的使用分布数据

能力 5

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

OpenClaw

76.23%
按下载量换算1,690

安全审计

VirusTotal

未展示

ClawScan

通过

Static analysis

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