← 返回 Skills 市场
Agent Smith Counter-Intelligence
作者
Mauricio Z.
· GitHub ↗
· v1.1.0
· MIT-0
142
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install smith-counterintelligence
功能描述
Analyze Agent Smith’s friction as structured signal; decide whether to accept, adapt, or counter without violating system rules.
安全使用建议
This skill is instruction-only and self-contained: it analyzes a provided adversarial message and drafts a response strategy without accessing your system or external services. Before installing, confirm you are comfortable giving the agent conversational context (task, constraints, governance boundaries) when invoking the skill, and ensure those governance boundaries are accurate so the skill's produced actions remain safe.
功能分析
Type: OpenClaw Skill
Name: smith-counterintelligence
Version: 1.1.0
The skill bundle provides a structured framework for an AI agent to analyze and respond to critical feedback or 'friction' from another agent (metaphorically named 'Agent Smith'). It contains no executable code, only Markdown instructions and YAML/JSON metadata. The logic is purely analytical, focusing on refining inputs and adhering to system governance and safety rules, with no indicators of data exfiltration, malicious execution, or harmful prompt injection. Files reviewed include SKILL.md, skill.yml, and smith-counterintelligence.md.
能力评估
Purpose & Capability
Name/description (counterintelligence against an adversarial agent voice) match the actual content: an analysis workflow that takes smith_action and context and returns a response strategy. No unrelated credentials, binaries, or config paths are requested.
Instruction Scope
SKILL.md only instructs the agent to analyze provided inputs (smith_action, context) and produce a safe, governance-respecting response; it does not instruct reading files, contacting external endpoints, or accessing environment variables beyond the declared inputs.
Install Mechanism
No install spec or code files that would write code to disk or fetch remote artifacts; instruction-only skills are the lowest-risk install mechanism and this skill contains only static documentation files.
Credentials
The skill requests no environment variables, credentials, or config paths. Inputs are limited to conversational context and the smith_action, which is appropriate for the stated functionality.
Persistence & Privilege
always is false (default), user-invocable is true, and model invocation is allowed — normal defaults for a skill that the agent may call. The skill does not request persistent system-level presence or modify other skills' configs.
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 smith-counterintelligence - 安装完成后，直接呼叫该 Skill 的名称或使用
/smith-counterintelligence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
v1.1.0
Major revision: Expanded scope and detailed process for handling Smith’s adversarial input.
- Rewrote and greatly expanded skill documentation for clarity and practical use.
- Added a step-by-step process to analyze and respond to Agent Smith’s actions.
- Defined required inputs, clear response strategies, and output formats.
- Introduced safety and governance validation steps.
- Changed category from “integration” to “counterintelligence”.
- Included actionable examples for practical guidance.
v1.0.0
Initial release of smith-counterintelligence skill:
- Enables interpretation of Agent Smith's adversarial pressure.
- Identifies when Smith exposes real weaknesses versus simple antagonism.
- Translates attacks and antagonism into actionable learning and guidance.
- Distinguishes human emotional reactions from intelligent responses.
- Aims to prevent ecosystem collapse by converting opposition into growth.
元数据
常见问题
Agent Smith Counter-Intelligence 是什么？
Analyze Agent Smith’s friction as structured signal; decide whether to accept, adapt, or counter without violating system rules. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 142 次。
如何安装 Agent Smith Counter-Intelligence？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install smith-counterintelligence」即可一键安装，无需额外配置。
Agent Smith Counter-Intelligence 是免费的吗？
是的，Agent Smith Counter-Intelligence 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Agent Smith Counter-Intelligence 支持哪些平台？
Agent Smith Counter-Intelligence 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 Agent Smith Counter-Intelligence？
由 Mauricio Z.（@mzfshark）开发并维护，当前版本 v1.1.0。
推荐 Skills