|
|
@@ -713,18 +713,22 @@
|
|
|
<div class="container">
|
|
|
<section class="section">
|
|
|
<div class="section-num">02</div>
|
|
|
- <h2 class="section-title">8维度<br>评估体系</h2>
|
|
|
- <p class="section-lead">100分评估体系。结构维度捕捉你能看到的问题,效果维度捕捉只有运行时才能感知的问题。</p>
|
|
|
+ <h2 class="section-title">9维度<br>评估体系</h2>
|
|
|
+ <p class="section-lead">100分评估体系。结构维度捕捉你能看到的问题,效果维度捕捉只有运行时才能感知的问题,Meta 维度检查「不要做什么」的反例编码。</p>
|
|
|
|
|
|
<div class="rubric-header">
|
|
|
<div class="rubric-stat">
|
|
|
- <div class="rubric-stat-num">60</div>
|
|
|
+ <div class="rubric-stat-num">59</div>
|
|
|
<div class="rubric-stat-label">结构<br>分值</div>
|
|
|
</div>
|
|
|
<div class="rubric-stat">
|
|
|
- <div class="rubric-stat-num rubric-stat-num--accent">40</div>
|
|
|
+ <div class="rubric-stat-num rubric-stat-num--accent">35</div>
|
|
|
<div class="rubric-stat-label">效果<br>分值</div>
|
|
|
</div>
|
|
|
+ <div class="rubric-stat">
|
|
|
+ <div class="rubric-stat-num">6</div>
|
|
|
+ <div class="rubric-stat-label">Meta<br>分值</div>
|
|
|
+ </div>
|
|
|
</div>
|
|
|
|
|
|
<table class="rubric-table">
|
|
|
@@ -741,37 +745,37 @@
|
|
|
<tr>
|
|
|
<td class="dim-num">1</td>
|
|
|
<td class="dim-name">Frontmatter质量</td>
|
|
|
- <td class="dim-weight">8</td>
|
|
|
- <td class="dim-desc">名称正确,描述包含功能/触发条件/使用场景,不超过1024字符</td>
|
|
|
+ <td class="dim-weight">7</td>
|
|
|
+ <td class="dim-desc">名称正确,描述包含功能/触发条件/使用场景,不超过1024字符,禁空话尾巴</td>
|
|
|
</tr>
|
|
|
<tr>
|
|
|
<td class="dim-num">2</td>
|
|
|
<td class="dim-name">工作流清晰度</td>
|
|
|
- <td class="dim-weight">15</td>
|
|
|
+ <td class="dim-weight">12</td>
|
|
|
<td class="dim-desc">步骤有编号、可执行,每步都有明确的输入/输出</td>
|
|
|
</tr>
|
|
|
<tr>
|
|
|
<td class="dim-num">3</td>
|
|
|
- <td class="dim-name">边界条件覆盖</td>
|
|
|
- <td class="dim-weight">10</td>
|
|
|
- <td class="dim-desc">错误处理、降级方案、常见故障恢复</td>
|
|
|
+ <td class="dim-name">失败模式编码</td>
|
|
|
+ <td class="dim-weight">12</td>
|
|
|
+ <td class="dim-desc">显式写出「如果 X 失败 → Y」的分支,有 fallback 路径与错误恢复</td>
|
|
|
</tr>
|
|
|
<tr>
|
|
|
<td class="dim-num">4</td>
|
|
|
<td class="dim-name">检查点设计</td>
|
|
|
- <td class="dim-weight">7</td>
|
|
|
- <td class="dim-desc">关键决策前需用户确认,防止自主失控</td>
|
|
|
+ <td class="dim-weight">6</td>
|
|
|
+ <td class="dim-desc">关键决策前需用户确认,检查点须显性标记(🔴/STOP/CHECKPOINT)</td>
|
|
|
</tr>
|
|
|
<tr>
|
|
|
<td class="dim-num">5</td>
|
|
|
- <td class="dim-name">指令具体性</td>
|
|
|
- <td class="dim-weight">15</td>
|
|
|
- <td class="dim-desc">无歧义,具体的参数/格式/示例,可直接执行</td>
|
|
|
+ <td class="dim-name">可执行具体性</td>
|
|
|
+ <td class="dim-weight">18</td>
|
|
|
+ <td class="dim-desc">无歧义,具体的参数/格式/示例,禁「视情况而定」类软化措辞</td>
|
|
|
</tr>
|
|
|
<tr>
|
|
|
<td class="dim-num">6</td>
|
|
|
<td class="dim-name">资源整合度</td>
|
|
|
- <td class="dim-weight">5</td>
|
|
|
+ <td class="dim-weight">4</td>
|
|
|
<td class="dim-desc">所有引用的脚本/资产路径存在且可访问</td>
|
|
|
</tr>
|
|
|
</tbody>
|
|
|
@@ -791,17 +795,37 @@
|
|
|
<tr>
|
|
|
<td class="dim-num">7</td>
|
|
|
<td class="dim-name">整体架构</td>
|
|
|
- <td class="dim-weight">15</td>
|
|
|
+ <td class="dim-weight">12</td>
|
|
|
<td class="dim-desc">层次清晰,无冗余或遗漏,符合生态系统约定</td>
|
|
|
</tr>
|
|
|
<tr>
|
|
|
<td class="dim-num">8</td>
|
|
|
<td class="dim-name">实测表现</td>
|
|
|
- <td class="dim-weight">25</td>
|
|
|
+ <td class="dim-weight">23</td>
|
|
|
<td class="dim-desc">运行2-3个测试提示词,对比启用 Skill 和 baseline 的输出质量</td>
|
|
|
</tr>
|
|
|
</tbody>
|
|
|
</table>
|
|
|
+
|
|
|
+ <table class="rubric-table">
|
|
|
+ <caption>Meta-skill 维度 — 反例与黑名单</caption>
|
|
|
+ <thead>
|
|
|
+ <tr>
|
|
|
+ <th style="width:36px">#</th>
|
|
|
+ <th style="width:180px">维度</th>
|
|
|
+ <th style="width:60px">权重</th>
|
|
|
+ <th>评分标准</th>
|
|
|
+ </tr>
|
|
|
+ </thead>
|
|
|
+ <tbody>
|
|
|
+ <tr>
|
|
|
+ <td class="dim-num">9</td>
|
|
|
+ <td class="dim-name">反例与黑名单</td>
|
|
|
+ <td class="dim-weight">6</td>
|
|
|
+ <td class="dim-desc">必须有「不要做什么」的反例清单,红灯/危险动作/反模式单独章节列出</td>
|
|
|
+ </tr>
|
|
|
+ </tbody>
|
|
|
+ </table>
|
|
|
</section>
|
|
|
</div>
|
|
|
|
|
|
@@ -872,9 +896,9 @@
|
|
|
</div>
|
|
|
<div class="phase-body">
|
|
|
<h3>Hill-Climbing 循环</h3>
|
|
|
- <p>按分数从低到高处理 Skill。每轮:诊断最弱维度,提出一个针对性修复,执行,重新评分,做出决定。</p>
|
|
|
+ <p>按分数从低到高处理 Skill。每轮:诊断加权短板最大的维度,提出一个针对性修复,执行,重新评分,做出决定。</p>
|
|
|
<ol class="phase-steps">
|
|
|
- <li>找出该 Skill 得分最低的维度</li>
|
|
|
+ <li>找出该 Skill 加权短板最大的维度(weight × (10 − score) / 10)</li>
|
|
|
<li>生成一项具体改进(改什么,为什么改,预期分数变化)</li>
|
|
|
<li>编辑 SKILL.md,用结构化消息 git commit</li>
|
|
|
<li>重新评分:结构由主 Agent,效果由独立子 Agent</li>
|
|
|
@@ -1009,13 +1033,13 @@
|
|
|
</tr>
|
|
|
<tr>
|
|
|
<td>评估指标</td>
|
|
|
- <td>8维度评估体系</td>
|
|
|
- <td>跨结构(60分)和效果(40分)的加权评分,总计100分。</td>
|
|
|
+ <td>9维度评估体系</td>
|
|
|
+ <td>跨结构(59分)、效果(35分)和 Meta(6分)的加权评分,总计100分。</td>
|
|
|
</tr>
|
|
|
<tr>
|
|
|
<td>实验循环</td>
|
|
|
<td>阶段2 hill-climbing</td>
|
|
|
- <td>诊断最弱维度,提出修复,执行,重新评分,保留或回滚。每个 Skill 最多3轮。</td>
|
|
|
+ <td>诊断加权短板最大的维度,提出修复,执行,重新评分,保留或回滚。每个 Skill 最多3轮。</td>
|
|
|
</tr>
|
|
|
<tr>
|
|
|
<td>版本控制</td>
|