Article optimizer
Skill xiaomoBoy/claude-writing-skills/skills/article-optimizer
Use when the user wants to improve a specific article under a fixed viral-scoring rubric, iterate on one article to raise its score, or run single-article scoring plus rewrite loops. Best for content optimization experiments where the scoring prompt stays fixed and the article text is the thing being changed.From its SKILL.md
npx -y skills add xiaomoBoy/claude-writing-skills --skill article-optimizerAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- runs commandsInstructs the agent to run 2 commands, including `python3 scripts/run_single_score.py <article_path>` and 1 more.
SKILL.md
2.2 KB, 573 tokens by cl100k_base, as published. Nobody here has run it
Article Optimizer
这个 skill 只负责一件事:在固定评分器下优化一篇具体文章。
它适合做的是:
- 给单篇文章打分
- 根据评分返回的改进建议改稿
- 做多轮版本迭代
- 对比不同版本的 viral score
它不负责:
- 调整评分器本身
- 校准整套评分系统
- 批量样本评估
如果目标是改评分 Prompt,而不是改文章,应该改用 score-optimizer。
Working Scope
适用场景:
- 用户说“把这篇文章优化到更高分”
- 用户说“按固定评分规则改稿”
- 用户说“给单篇文章打分并根据建议改写”
- 用户说“跑 single article optimization loop”
不适用场景:
- 用户说“让这套评分系统更准”
- 用户说“批量评估一组样本文章”
Required Reads
开始前先读:
需要看实现时再读:
Workflow
- 先确认当前任务是在改文章,不是在改评分器。
- 读取
references/optimize_program.md,确认实验约束。 - 确认目标文章路径。
- 用单篇打分脚本跑出基线分数和改进建议。
- 每轮只改文章,不动评分规则。
- 保存迭代版本,再重新打分比较。
Commands
在 skill 根目录执行:
python3 scripts/run_single_score.py <article_path>
如果需要图表:
python3 scripts/generate_optimize_chart.py
Guardrails
- 默认不要改
references/score_prompt.md - 默认不要改
scripts/run_single_score.py - 每轮优先只改文章内容
- 如果用户要的是“训练裁判”,不要用这个 skill
What ships with it: 5 files
16.0 KB alongside SKILL.md, 2 of them executable
references/
- optimize_program.md3.3 KB
- score_prompt.md3.8 KB
scripts/
- generate_optimize_chart.pyruns3.7 KB
- run_single_score.pyruns3.6 KB
- README.md1.6 KB