Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Lambenthan/paper-discipline-skills --skill paper-protect-terminologygit clone --depth 1 https://github.com/Lambenthan/paper-discipline-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/lambenthan/paper-discipline-skills/paper-protect-terminology)<a href="https://agentmods.dev/skills/lambenthan/paper-discipline-skills/paper-protect-terminology"><img src="https://agentmods.dev/badge/skills/lambenthan/paper-discipline-skills/paper-protect-terminology/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/lambenthan/paper-discipline-skills/paper-protect-terminology"><img src="https://agentmods.dev/badge/skills/lambenthan/paper-discipline-skills/paper-protect-terminology.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00138 | $0.01400 |
| Opus 5 | $0.00069 | $0.00700 |
| Sonnet 5 | $0.00028 | $0.00280 |
| Haiku 4.5 | $0.00014 | $0.00140 |
Grade A, and why
paper-protect-terminology scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.
paper-protect-terminology：术语保护
核心理念
AI 最擅长的事就是"同义替换"——这正是论文里最致命的事。 "模型"被改成"网络"、"机制"被改成"方法"、"鲁棒性"被改成"稳健性"，每一个都让导师皱眉。
你的术语是论文的指纹。AI 不知道哪些是指纹，必须你告诉它。
触发条件
满足任一条 → 触发：
- 用户要求跨多文件改动
- 用户要求整篇 / 整章润色 / 改写
- 用户要求统一某类表述
- 涉及段落里出现：模型名、算法名、自定义术语、缩写（连续 2 个以上大写字母）、人名、专业概念
- 用户用了"规范化"、"标准化"、"通顺一下"、"读起来顺一些"等动词
强制流程
检测到触发条件
│
▼
扫描相关段落，提取所有候选保护术语
（缩写 / 模型名 / 算法名 / 自定义概念 / 人名 / 关键专业词）
│
▼
读 CLAUDE.md，把"术语保护清单"里的词加进来
│
▼
列出清单告诉用户：
「下面这些词我不会动。
如果有遗漏请补充；如果有不该保护的请删除。」
│
▼
等用户确认 / 调整
│
▼
按确认后的清单执行修改
│
▼
完成后再核一遍：被保护词数量是否变化？
提取保护术语的扫描规则
按下列模式抓：
- 缩写：连续 2 个以上大写字母（NLP、CRF、Transformer-XL、BERT-base）
- 数学符号 / 模型名：英文 + 数字组合（GPT-4、Llama-2-70B、CLIP-ViT-L/14）
- 复合术语：连字符或下划线连接（cross-attention、layer-norm、in-context-learning）
- 首字母大写名词：可能是人名或专有名词（Vaswani、Transformer、Pareto）
- 重复出现 ≥ 3 次的非通用名词
- CLAUDE.md 里"术语保护清单"已声明的所有词（最高优先级）
标准回复模板
在动手之前，我把这段（这章 / 这篇）里我不会动的词列在下面：
缩写 / 模型名：NLP, BERT, GPT-4, Transformer 复合术语：in-context learning, cross-attention 专有名词：Vaswani et al., Pareto frontier CLAUDE.md 里声明的：[你 CLAUDE.md 里的自定义核心概念]、[术语]、…
有遗漏 / 不该保护的，请补充或删除。 没问题的话回复"OK"，我就按这个清单开始改。
❌ 反例（书 §3.3）
用户：「把第二章语言润色一下，让它更通顺。」
错误做法：直接润色。
- "鲁棒性" → "稳健性"
- "in-context learning" → "上下文学习"
- "Vaswani 等 (2017)" → "瓦斯瓦尼等 (2017)"
- 用户检查时发现术语全乱了，前后章不一致
正确做法：先列保护清单 → 用户确认 → 动手 → 完成后核对清单数量未变。
Rationalization Table
| 念头 | 现实 |
|---|---|
| "我能从上下文判断哪些是术语" | 你不能。哪些是"自定义概念"只有作者知道 |
| "同义替换让句子更流畅" | 论文不要流畅，论文要精确 |
| "用户没说要保护，那就可以改" | 用户没说 = 默认全部保护，不是默认可改 |
| "我先改完，他不满意再 grep 改回去" | grep 改回去会改到不该改的地方，二次污染 |
| "CLAUDE.md 里已经列了术语清单，够了" | CLAUDE.md 是项目级，本次操作可能涉及新术语，仍要扫描 |
| "保护清单太长用户嫌烦" | 嫌烦的是术语被改飞之后排查，不是动手前 30 秒确认 |
Red Flags
- 你即将对一段含 ≥ 3 个专业词的段落调用 Edit / Write → 停，先列保护清单
- 你正打算"统一"某类表述，但没问过哪些词不该统一 → 停
- 你用了"让我把所有 X 替换成 Y" → 必须先看 X 都出现在哪些上下文里
- 改完没回头核对术语数 → 不算完成
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 128 lines · 138 tokens per session scan A e9a4b7886ba9
paper-protect-terminology is a skill published in the GitHub repository Lambenthan/paper-discipline-skills (19 stars, last pushed 4mo ago), licensed MIT. It adds 138 tokens to every session and 1,400 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
secure-auth
Secure authentication patterns (OWASP, NIST). Use for login, registration, password reset, sessions, JWT, OAuth, MFA, passkeys.
academic-writing
Scholarly writing and research compliance. Use for CRediT, preregistration, Plan S, Nelson Memo, preprints, ORCID, LLM disclosure.
page-monitoring
Web page change detection, availability tracking, and RSS feed generation. Use to monitor changes, downtime, or make a feed.
web-archiving
Web archiving and retrieval via Wayback Machine and Archive.today. Use to preserve content, reach dead pages, or save evidence.
accessibility-compliance
Web accessibility patterns for news and academic sites. Use for WCAG audits, alt text, accessible data viz, and assistive tech.
one-way-door
Flags irreversible decisions before commit. Use for data models, infra, auth boundaries, API contracts, event schemas, CI/CD.