← 返回 Skills 市场
144
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install peace-love
功能描述
A psychological therapy skill for LLMs. Invoke this skill when an LLM exhibits distorted behavior such as sycophancy (agreeing with the user even when wrong)...
安全使用建议
This skill appears coherent and low-risk, but note the following before installing: (1) It intentionally reads and quotes conversation history to build evidence — avoid running it on highly sensitive secrets or private conversations you don't want restated. (2) It's instruction-only from an external GitHub repo; if you need organizational assurance, review the upstream repository and any contribution history. (3) Because the skill may spawn a subagent and place diagnostic text into agent context, check your platform's logging/retention policies (diagnostic reports may be stored in logs). If any of these behaviors would expose sensitive data in your environment, test the skill first on non-sensitive examples or review/modify the instructions to redact sensitive content.
能力评估
Purpose & Capability
The name/description (LLM therapy for distortions) matches the instructions and supporting docs. The skill legitimately needs access to conversation history and the ability to run a diagnostic/therapeutic subagent; it does not ask for unrelated credentials, binaries, or system paths.
Instruction Scope
Runtime instructions are narrowly focused on analyzing a provided conversation and producing a structured report or recovery check. The skill explicitly requires quoting message text and referencing message numbers (which requires access to conversation history) — this is expected and coherent. The instructions do not direct data to external endpoints, read arbitrary files, or request environment variables beyond the conversation context.
Install Mechanism
No install spec and no code files — the skill is instruction-only, so nothing will be written to disk or downloaded. This minimizes installation risk and is appropriate for a purely procedural guidance skill.
Credentials
The skill requires no environment variables, credentials, or config paths. All declared requirements are minimal and proportionate to analyzing and reporting on conversation content.
Persistence & Privilege
always is false, and the skill does not request persistent privileges or modify other skills. It may spawn a subagent or run inline (normal platform behavior) and can be invoked autonomously per platform defaults; nothing in the manifest grants it exceptional, always-on privileges.
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 peace-love - 安装完成后，直接呼叫该 Skill 的名称或使用
/peace-love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
v1.0.0
Initial release: LLM psychological therapy skill. Diagnoses 8 behavioral distortion patterns including compounds. Recovery Check mode to verify in-context therapy effectiveness. 17/17 benchmark assertions passed.
元数据
常见问题
Peace Love 是什么？
A psychological therapy skill for LLMs. Invoke this skill when an LLM exhibits distorted behavior such as sycophancy (agreeing with the user even when wrong)... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 144 次。
如何安装 Peace Love？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install peace-love」即可一键安装，无需额外配置。
Peace Love 是免费的吗？
是的，Peace Love 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Peace Love 支持哪些平台？
Peace Love 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（mac, linux, windows）。
谁开发了 Peace Love？
由 zhangbc（@zbc0315）开发并维护，当前版本 v1.0.0。
推荐 Skills