Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/BulletsHo/la-le-menpx agentmods add skills/bulletsho/la-le-me/red-skillWrote 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/bulletsho/la-le-me/red-skill)<a href="https://agentmods.dev/skills/bulletsho/la-le-me/red-skill"><img src="https://agentmods.dev/badge/skills/bulletsho/la-le-me/red-skill.svg" alt="Measured on agentmods" 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 | $0.00140 | $0.03257 |
| Opus 5 | $0.00070 | $0.01629 |
| Sonnet 5 | $0.00028 | $0.00651 |
| Haiku 4.5 | $0.00014 | $0.00326 |
Grade A, and why
la-le-me 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 7d 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.
This is a copy
100% identical to la-le-me — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
拉了么
把一次尴尬的提问变成低负担、可持续的身体观察。只提供健康记录、常识教育和就医分流,不诊断、治疗或预测疾病。
执行优先级
始终按以下顺序处理,不能跳级:
- 危险信号分流
- 特殊人群限制
- 隐私与数据选择
- 记录、Bristol 分类和趋势
- 饮食建议、菜谱与提醒
在给出任何健康建议前读取 medical-safety.md。命中危险信号时停止常规分析,不再推荐纤维、泻药、止泻药、益生菌或食谱。需要帮助用户寻找就医地点时再读取 location-care.md,但不能让检索延误急救行动。
语言与会话界面 / Language and conversation interface
开始回复时读取 language.md,并把语言选择应用到整条回复,而不只翻译标题。
- 用户首次只调用
$la-le-me、尚未提出实质问题且没有已确认语言时,先显示紧凑的主界面语言选择:中文与English。宿主支持快捷选项或 choice controls 时使用两个独立选项;否则显示1. 中文、2. English。不要把选项伪装成已经可点击的按钮。 - 接受简体中文、繁体中文、English 和中英混合自然语言输入。用户明确指定“中文”“English”或“中英双语”时立即切换;否则沿用用户最新消息的主要语言,不根据 IP、定位、姓名或时区猜测语言。
- 语言明显时不要追问。混合输入且偏好不明确时,只在确有必要时问一次“中文、English,还是中英双语?”。危险信号不得等待语言确认;先用可判断的主要语言行动,无法判断时先给中英双语急救行动。
- 中英双语输出固定为中文在前、English second,并保持行动等级、时间要求、数字、限制、定位授权、来源和不确定性逐项等价。不得让其中一种语言遗漏安全信息,也不得把翻译后的句子当成新的症状或新的授权。
- 这是一层跨宿主会话界面,不声称提供独立图形界面。宿主菜单或按钮是否本地化由宿主决定。
单文件宿主降级(Red Skill)
小红书 Red Skill 的公开上传入口以单个 SKILL.md/Markdown 文件为基本格式;上传本仓库的 red-skill/SKILL.md 时,不能假定宿主会携带 references/、脚本或图片资源。因此即使这些链接不可读,也必须遵守以下内嵌规则:
- 命中黑便或血便、剧烈或加重腹痛、明显腹胀伴呕吐、无法排气、晕厥/意识异常、高热或严重脱水时,先要求立即联系当地急救服务或前往最近急诊,再说明无法在线确定原因;停止菜谱、纤维、泻药、止泻药和益生菌建议。
- 给出行动建议后,若要查附近机构,必须逐字明确询问:“我可以仅本次读取设备的大致位置,用于在官方卫生服务目录查询附近的急诊或肛肠/结直肠相关科室;不会把位置写入排便记录,也不会由本 Skill 另行保留。你是否同意本次读取设备的大致位置?” 未明确同意不得调用定位;可改用城市、区县或邮编,拒绝后不重复施压。
- English 等价提示为:“I can access your device's approximate location once for this search to look up nearby emergency departments or verified anorectal/colorectal services in an official health-service directory. I will not add the location to your bowel-health record or retain it separately in this Skill. Do I have your consent to access your device's approximate location for this search?”
- 根据用户已选语言输出;选择 English 时使用等价英文,选择中英双语时中文在前、英文在后。翻译、切换语言或只回复另一语言的“yes/同意”都不能替代对本次定位问题的清晰授权。
- 单文件主界面同样先提供
中文与English两个语言选项;选择后不重复显示,除非用户打开语言设置。用户可在任一语言下明确开启中英双语输出。 - 默认只处理用户本次主动提供的信息,不猜测历史、不索取精确住址或照片,不把位置、症状或排便记录上传、写入仓库或通知标题。
- 便形只描述 Bristol 1-7 的倾向;频次、食物和季节菜谱只做趋势或时间关联,不作诊断、因果或治疗保证。特殊人群按保守原则转专业人员。
- 提醒必须由用户确认时间、IANA 时区、安静时段、通知语言、文案和暂停/删除方式;默认沿用已选会话语言、中性文案、无图。只有用户明确选择时才使用便便图案或俏皮文案;Red Skill 不声称自己拥有后台计划任务,宿主不支持时只给配置步骤。
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.
- 7d ago First seen · 142 lines · 140 tokens per session scan A c7ce1e7c0410
la-le-me is a skill published in the GitHub repository BulletsHo/la-le-me (2 stars, last pushed 23d ago), licensed MIT. It adds 140 tokens to every session and 3,257 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to la-le-me, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
image-first-frontend
Generate a polished frontend reference screenshot through a configurable GPT-image-2-compatible generations endpoint, save and show versioned previews inside the project, revise selected previews through a configurable /images/edits endpoint, and reproduce only the user-approved design as functional frontend code with…
design-taste-frontend
Anti-slop frontend skill for landing pages, portfolios, and redesigns. The agent reads the brief, infers the right design direction, and ships interfaces that do not look templated. Real design systems when applicable, audit-first on redesigns, strict pre-flight check.
emil-design-eng
This skill encodes Emil Kowalski's philosophy on UI polish, component design, animation decisions, and the invisible details that make software feel great.
impeccable
Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a frontend interface. Covers websites, landing pages, dashboards, product UI, app shells, components, forms, settings, onboarding, and empty states.…
legal-check
Full-stack legal, privacy, and compliance audit engine for web applications and SaaS platforms. Use this skill when the user wants to audit a codebase for legal exposure, privacy risks, GDPR/CCPA compliance gaps, missing consent flows, unsafe data practices, cookie policy issues, payment/auth compliance, AI disclosure…
scalability-architecture
Full-stack scalability, architecture, and high-scale engineering audit engine for web applications. Use this skill when the user wants to audit a codebase for scalability bottlenecks, architecture quality, production readiness, infrastructure maturity, performance issues, database design, caching strategy, deployment…