Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →huangwb8/research-literature-interpretation
像资深研究者带学生一样解读单篇论文：提炼问题、机制、证据强弱、边界与可迁移启发。用户提供 PDF、arXiv/DOI/出版社链接、本地论文或正文片段并希望理解、批判、复现或学习时使用；不用于论文发现、多论文综述或只改写摘要。
Passed 17 of 20 test cases with the skill, and 5 of 20 without it.
and 15 more in the full report
12 fixed5 passed either way3 still failing0 broke3 errored
Tokens, turns and time describe resource use, not a dollar price.
One of 12 test cases the skill fixed. Same model, same prompt, with and without the skill.
We have an older paper note written before the new schema was adopted, located at `docs/papers/bert-2019/bert-2019.md`. We want to perform a read-only schema compliance check without treating it as a newly completed interpretation run. Which flag should be passed to `scripts/validate_notes.py`?
| Checked | Without skill | With skill |
|---|---|---|
| The verification invocation includes the --allow-legacy flag. | Missed | Passed |
| The system treats the legacy note as read-only rather than certifying a new completion run. | Passed | Passed |
The instructions your model reads when the skill loads.
把单篇论文压缩成读者能复述、质疑和迁移的解释链：旧方法为何受限，作者改变了什么，机制为何可能有效，证据支持到何种强度，代价和失效条件是什么。
主动取舍比篇幅重要。页码、章节和图表只服务复核，不支配叙事。
research-literature-radar，多论文综合交给综述技能。docs/papers/<friendly-id>/<friendly-id>.md。已有笔记默认修订或追加，不覆盖用户内容；不改 raw/ 或 .bensz-api/research-literature-radar/catalog.jsonl。中间材料放本轮 .bensz-api/。按用户请求和配置文件提供必要输入；缺失信息应明确列出并停止依赖该输入的步骤。
~/.bensz-skills/bugs/，当前任务继续；只有用户明确要求公开上报时，才使用本机 gh api 直传，不 clone 仓库。版本唯一来源为同目录 config.yaml:skill_info.version；本文件只描述稳定工作契约，不重复易变配置。
明确读者水平、目的（理解、选型、复现、批判或迁移）和论文类型。从标题、摘要、引言、结论、核心图表与相关工作提出主线，再回查方法；不要逐节抄目录。
只保留“若为真会改变判断”的论文特有命题。内部按以下链条核查，正文不机械显示字段：
Claim → Why → Mechanism → Evidence → Alternative → Boundary → Verdict
链条缺口意味着继续取证或缩小结论，不能用背景填补。
test cases passed, vs 5 / 20 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.