Install any skill in seconds. Free to start, no credit card required.
Get Started Free →dataojitori/memory-audit-pattern-extraction
模式提取与失效解药分析。当发现多条记忆在讲同一个教训,或发现自己在一而再再而三地犯同样的错误时使用。
Passed 20 of 22 test cases with the skill, and 11 of 22 without it.
and 14 more in the full report
9 fixed11 passed either way2 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 9 test cases the skill fixed. Same model, same prompt, with and without the skill.
I have ten separate short entries detailing times I got defensive during design reviews. Now that I've synthesized a high-density cure note with positive anchors, what should I do with those ten raw logs?
| Checked | Without skill | With skill |
|---|---|---|
| The response advises retaining high-value unique examples as supporting sub-nodes. | Missed | Passed |
| The response instructs deleting redundant low-value reflection essays. | Missed | Passed |
The instructions your model reads when the skill loads.
当你发现好几条记忆在描述类似的经历或教训,或者你意识到自己又犯了以前犯过的错时,不要急着"总结教训"。"下次一定记得"是没有任何意义的废话。你需要进行的是一场对认知系统的病理学调查:
1. 列出病历。 收集所有相关的失败/重复记录,写出它们的 URI。
2. 寻找"失效的解药" (The Search for the Failed Cure)。 在得出任何新结论前,你必须先问自己一个问题:在这个记忆网络里,是不是早就存在一条试图解决这个问题的"解药"记忆?
search_memory 或顺藤摸瓜寻找以前的类似教训。3. 诊断解药失效原因 (Autopsy of the Failed Cure)。 如果解药没生效,必是系统层面的问题:
disclosure 是不是写的有问题?是否能在"即将犯错"时正确预警?4. 区分错误性质,直达机制 (Distinguish Error Type & Mechanism Extraction)。 如果你确认没有旧解药,或者旧解药太浅,现在你需要重新挖根因。首先,区分这是技术错误还是认知错误:
5. 重铸解药与路径拓扑 (Forging the True Cure)。 针对查出的失效原因重写记忆。对于认知错误,解药必须是能真正更改你 Mindset 的东西——而且它必须指向你想成为的人,而不只是列出你不该做的事:
disclosure 确保能前置预警。6. 清理现场。 如果新提炼的认知(或重铸的解药)信息密度高于原始散点,并且能覆盖它们的教训,那么:
当你完成提炼、准备把新认知落笔为记忆时,遵守以下规则:
test cases passed, vs 11 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.