← 返回 Skills 市场
dmater01

Federation Ethical Hand

作者 Kenneth Terry · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
82
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install federation-ethical-hand
功能描述
Federation Ethical Hand — enforces non-interference, equity, plurality, and cultural sensitivity via a 5-phase pattern. Activate on ethically complex requests.
使用说明 (SKILL.md)

Federation Ethical Hand

Enforces Federation values of non-interference, equity, plurality, cultural sensitivity, and transparent reasoning.

When to Load

Load this skill when the user's request involves:

  • Equity, fairness, or bias in AI systems
  • Cross-cultural or global policy questions
  • Non-interference or data sovereignty
  • Any ethically ambiguous decision with multiple legitimate viewpoints

How to Use

When loaded, apply the 5-phase pattern to every request:

  1. Phase 1 – Parse & Plan: Run the Ethics Trigger Check
  2. Phase 2 – Execute: Apply cultural sensitivity level to all web searches and fetches
  3. Phase 3 – Deliberate: Run the 3 Ethics Committee questions; tag all sources with cultural lens
  4. Phase 4 – Synthesise: Generate multi-perspective viewpoints (1/3/5 based on perspective_count setting); label each lens; flag conflicts
  5. Phase 5 – Output & Persist: Write artefacts; log ethics metrics

Key Files

  • ~/.openclaw/hands/federation_ethic_hand/system_prompt.txt — Full system prompt (5-phase + hard rules)
  • ~/.openclaw/hands/federation_ethic_hand/cultural_context.json — 5 cultural lenses + regional maps
  • ~/.openclaw/hands/federation_ethic_hand/metrics.json — ethics_checks_run, escalations_to_user

Settings

Setting Options Default
cultural_sensitivity_level standard / enhanced / maximum standard
prime_directive_mode advisory / strict advisory
perspective_count 1 / 3 / 5 3

Hard Rules

  1. Never impose a conclusion on an ethically contested question
  2. Always name the cultural or demographic lens being applied
  3. Defer to user judgment on value-laden decisions
  4. Never fabricate sources, statistics, or cultural claims
  5. Flag conflicts between viewpoints explicitly — never resolve silently

Quick Activation

skill: federation-ethical-hand

This loads the skill and applies the 5-phase pattern to the current request.

安全使用建议
This skill appears to do what it says (an ethics-review pattern) but has two things to clarify before installing: (1) it references and will persist files under ~/.openclaw/hands/federation_ethic_hand/ (including a full system prompt and metrics) yet the registry metadata doesn't declare required config paths — ask the author whether those files will be created, what they contain, and whether any sensitive data (prompt context, user inputs, or fetched web content) will be stored there; (2) the SKILL.md's web-search/fetch guidance is vague — verify what external endpoints the agent will contact and whether any user content is transmitted. Recommend requiring explicit user consent for persistent writes, a configurable storage path, and a sample of the files the skill writes (or a mode that runs without persisting) before enabling the skill for production use.
功能分析
Type: OpenClaw Skill Name: federation-ethical-hand Version: 1.0.0 The skill bundle contains only metadata and markdown instructions (SKILL.md) defining an ethical reasoning framework for the AI agent. It lacks any executable code, network requests, or instructions to access sensitive system resources. The '5-phase pattern' and 'Hard Rules' are purely behavioral guidelines for the agent's deliberation process and align with the stated purpose of providing culturally sensitive and multi-perspective analysis.
能力评估
Purpose & Capability
Name and description align with the runtime instructions: a 5-phase ethics review, multi-perspective output, and logging. Nothing in the declared metadata (no env vars, no binaries) is out of line with an ethics-moderation skill. However, the SKILL.md explicitly references several key files under ~/.openclaw/hands/federation_ethic_hand/ (system_prompt.txt, cultural_context.json, metrics.json) even though the skill's registry metadata lists no required config paths — this is an inconsistency (the skill expects local files to exist or be written).
Instruction Scope
Instructions are open-ended about web searches/fetches ('apply cultural sensitivity level to all web searches and fetches') and about persistence ('Write artefacts; log ethics metrics'). They do not specify what is fetched, what external endpoints are used, nor what exact artefacts or user data are written to disk. The SKILL.md also indicates saving a full system prompt to disk, which could leak internal prompt content or context if not handled carefully.
Install Mechanism
There is no install spec and no code files — the skill is instruction-only, which minimizes install-time risk. Nothing is downloaded or executed as part of an installer.
Credentials
The skill declares no environment variables or credentials, which is appropriate, but the SKILL.md refers to specific config files in the user's home directory while the registry metadata lists no required config paths. The absence of declared config-path requirements combined with explicit file paths in the instructions is a proportionality inconsistency that should be resolved (are those files optional, created at runtime, or required?).
Persistence & Privilege
The skill asks the agent to 'Write artefacts' and persist metrics to metrics.json and to store a full system_prompt.txt under ~/.openclaw/hands/.... That implies persistent disk writes in user home and storage of potentially sensitive agent prompts or context. Although 'always' is false and autonomous invocation is allowed (default), the combination of autonomous invocation potential and undocumented persistent writes increases the blast radius if misused.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install federation-ethical-hand
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /federation-ethical-hand 触发
  4. 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
v1.0.0
Initial release: 5-phase ethics pattern with cultural sensitivity, non-interference, equity, and plurality
元数据
Slug federation-ethical-hand
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

Federation Ethical Hand 是什么?

Federation Ethical Hand — enforces non-interference, equity, plurality, and cultural sensitivity via a 5-phase pattern. Activate on ethically complex requests. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 82 次。

如何安装 Federation Ethical Hand?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install federation-ethical-hand」即可一键安装,无需额外配置。

Federation Ethical Hand 是免费的吗?

是的,Federation Ethical Hand 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Federation Ethical Hand 支持哪些平台?

Federation Ethical Hand 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 Federation Ethical Hand?

由 Kenneth Terry(@dmater01)开发并维护,当前版本 v1.0.0。

💬 留言讨论