← 返回 Skills 市场
135
总下载
0
收藏
0
当前安装
5
版本数
在 OpenClaw 中安装
/install pm-article-scorer
功能描述
AI产品经理内容相关性评分技能。对公众号文章进行多维度启发式评分（0-100分），判断是否值得AI产品经理关注。
安全使用建议
This skill is an instruction-only heuristic rubric and appears coherent and low-risk. Before installing/using, verify the actual implementation of heuristic_score will: (1) run locally without calling external APIs or sending content offsite, (2) avoid logging or transmitting article text or metadata to third parties, and (3) match the rubric's output format. Be aware the SKILL.md biases scoring toward certain named '优质原创作者' — consider whether that weighting is acceptable. If you let an assistant implement the function, review its code/output to ensure no hidden network calls or credential usage.
功能分析
Type: OpenClaw Skill
Name: pm-article-scorer
Version: 1.0.4
The skill bundle contains instructions for an AI agent to perform heuristic scoring of articles for AI Product Managers. The logic is entirely contained within SKILL.md as a scoring rubric and does not include any executable code, network requests, or data exfiltration mechanisms. The metadata and instructions are consistent with the stated purpose of content filtering.
能力评估
Purpose & Capability
Name/description (PM article scoring) match the content of SKILL.md. No unrelated env vars, binaries, or config paths are requested.
Instruction Scope
SKILL.md defines detailed local heuristic rules and an expected return dict. It states processing is local and no external API/LLM calls are required. However, the skill is instruction-only and provides no implementation code — the agent or integrator must implement heuristic_score(title, author, content), so actual runtime behavior depends on that implementation.
Install Mechanism
No install spec and no code files — lowest-risk form. Nothing is downloaded or written to disk by the skill package itself.
Credentials
No environment variables, credentials, or config paths are requested. Declared requirements are minimal and appropriate for the stated purpose.
Persistence & Privilege
Skill is not marked always:true and is user-invocable. It does not request persistent or privileged system presence.
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 pm-article-scorer - 安装完成后，直接呼叫该 Skill 的名称或使用
/pm-article-scorer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
v1.0.4
清理遗留文件（移除README.md、旧版scripts、OpenAI API key示例），仅保留SKILL.md+LICENSE+_meta.json
v1.0.3
修复_meta.json错误元数据（ownerId/slug指向前一版本的humanizer）；更新评分为语义价值判断逻辑
v1.0.2
评分逻辑全面重构：从关键词计数升级为语义价值判断，新增内容类型判断、个人输出加权、营销号降权等修正机制
v1.0.1
娣诲姞 footer锛氭杩庡叧娉ㄤ綔鑰呭井淇″叕浼楀彿锛氬紑璁板仛浜у搧
v1.0.0
Initial release: AI产品经理内容相关性评分工具，支持多维度评分（0-100分），判断公众号文章是否值得阅读
元数据
常见问题
Pm Article Scorer 是什么？
AI产品经理内容相关性评分技能。对公众号文章进行多维度启发式评分（0-100分），判断是否值得AI产品经理关注。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 135 次。
如何安装 Pm Article Scorer？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install pm-article-scorer」即可一键安装，无需额外配置。
Pm Article Scorer 是免费的吗？
是的，Pm Article Scorer 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Pm Article Scorer 支持哪些平台？
Pm Article Scorer 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 Pm Article Scorer？
由 Kai（@agasding）开发并维护，当前版本 v1.0.4。
推荐 Skills