About score-optimizer
A skill for tuning article-scoring systems, intended for content teams calibrating viral scoring prompts and rules. It scores sample collections against labeled data and evaluates results; by default, it changes the scoring prompt rather than article content.
Key features
- 01Adjust article-scoring prompts
- 02Batch-score sample collections
- 03Evaluate accuracy against labeled data
- 04Compare MAE, Spearman, and other metrics
- 05Avoid changing evaluation scripts by default
Use cases
- 01Improve a viral-scoring prompt
- 02Calibrate viral score or composite score
- 03Run a sample-collection scoring experiment
- 04Check scoring rules against labeled results
Install
Copy the command and run it in a terminal.
$
npx skills add https:/ / github.com/ xiaomoboy/ claude-writing-skills --skill score-optimizerAfter installing, type/score-optimizerin Claude Code to confirm it loaded.
Bundled files
References and scripts installed along with SKILL.md.
README.mdassets/articles/labels.example.jsonassets/articles/samples/README.mdreferences/score_program.mdscore_prompt.mdscripts/evaluate.pygenerate_score_chart.pyrun_scoring.pyDetails
From the project READMEScore Optimizer
这个 skill 只负责一件事:优化文章评分系统本身。
它适合做的是:
- 调整文章爆款评分 Prompt
- 批量给样本文集打分
- 对照标注数据评估评分准确度
- 研究评分维度、权重、few-shot 示例是否更合理
它不负责:
- 直接改文章正文
- 平台改写
- 发布包装
Working Scope
适用场景:
- 用户说“优化文章评分 prompt”
- 用户说“让这个爆款评分系统更准”
- 用户说“跑一轮样本文集评分实验”
- 用户说“校准 viral score / composite score 的打分规则”
不适用场景:
- 用户说“直接把这篇文章改高分”
- 用户说“按这个评分器去优化文章内容”
后者应改用 article-optimizer。
Required Reads
开始前先读:
需要看实现时再读:
样本文集位置(用户自己提供,本 skill 不带样本):
assets/articles/samples/—— 把你自己标注过的 .md 文章放进来,然后cp labels.example.json labels.json填上你的评分。详见 assets/articles/samples/README.md
Workflow
- 先确认当前任务是在优化评分器,不是在改文章。
- 读取
references/score_program.md,按它的实验约束执行。 - 默认只改
references/score_prompt.md,不要动评估脚本。 - 运行批量评分脚本,再运行评估脚本。
- 比较
composite_score、mae、spearman、classification。 - 记录结果,决定保留还是回滚。
Commands
在 skill 根目录执行:
python3 scripts/run_scoring.py > scoring.log 2>&1
python3 scripts/evaluate.py > eval.log 2>&1
grep "^composite_score:\|^mae:\|^spearman:\|^classification:" eval.log
如果需要图表:
python3 scripts/generate_score_chart.py
Guardrails
- 默认不要改
scripts/evaluate.py - 默认不要改
scripts/run_scoring.py - 默认不要安装新依赖
- 如果只是想提高某一篇文章分数,不要用这个 skill
Comments
Markdown supported