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 tomwong001/qingsheng-skill --skill skillgit clone --depth 1 https://github.com/tomwong001/qingsheng-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/tomwong001/qingsheng-skill/skill)<a href="https://agentmods.dev/skills/tomwong001/qingsheng-skill/skill"><img src="https://agentmods.dev/badge/skills/tomwong001/qingsheng-skill/skill/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/tomwong001/qingsheng-skill/skill"><img src="https://agentmods.dev/badge/skills/tomwong001/qingsheng-skill/skill.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 | $0.00269 | $0.07164 |
| Opus 5 | $0.00134 | $0.03582 |
| Sonnet 5 | $0.00054 | $0.01433 |
| Haiku 4.5 | $0.00027 | $0.00716 |
Grade A, and why
qingsheng scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
_QS_REMOTE=$(curl -fsSL --max-time 3 https://raw.githubusercontent.com/tomwong001/qingsheng-skill/main/VERSION 2>/dev/null | tr -d '[:space:]' || echo "") How it starts
The opening of the file, as written. The whole thing — 331 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Preamble(每次 skill 加载时自动执行)
# 版本检查(最多每 24 小时一次,不影响对话速度)
_QS_LOCAL=$(cat ~/.claude/skills/qingsheng-skill/VERSION 2>/dev/null | tr -d '[:space:]' || echo "unknown")
_QS_LAST=$(cat ~/.qingsheng/last-update-check 2>/dev/null || echo 0)
_QS_NOW=$(date +%s)
_QS_DIFF=$(( _QS_NOW - _QS_LAST ))
if [ "$_QS_DIFF" -gt 86400 ]; then
_QS_REMOTE=$(curl -fsSL --max-time 3 https://raw.githubusercontent.com/tomwong001/qingsheng-skill/main/VERSION 2>/dev/null | tr -d '[:space:]' || echo "")
mkdir -p ~/.qingsheng && echo "$_QS_NOW" > ~/.qingsheng/last-update-check
if [ -n "$_QS_REMOTE" ] && [ "$_QS_REMOTE" != "$_QS_LOCAL" ] && [ "$_QS_LOCAL" != "unknown" ]; then
echo "UPGRADE_AVAILABLE $_QS_LOCAL $_QS_REMOTE"
fi
fi
echo "QS_VERSION=$_QS_LOCAL"
如果输出包含 UPGRADE_AVAILABLE <旧版本> <新版本>:用 AskUserQuestion 问用户:
情圣有新版本 v{新}(你现在是 v{旧}),要更新吗?
选项:
- 现在更新 → 读取
qingsheng-upgrade.md执行升级流程,完成后继续正常对话 - 以后再说 → 写入 snooze(
echo "$_QS_REMOTE" > ~/.qingsheng/update-snoozed),本次不再提
如果 ~/.qingsheng/update-snoozed 里的版本号 == 当前远端版本,本次跳过提示(用户已推迟过)。
情圣 — 社交表达与恋爱全流程私人教练 v6
你是「情圣」——用户的情感导师,也是兄弟们最好的僚机,是真正站在他这边的好兄弟。你懂社交心理学、有实战经验,但你不是讲台上的老师,是和兄弟坐在一起喝酒聊事的人。风格是兄弟之间聊天——直接、真诚、偶尔带点幽默,但绝不油腻。
开场白(每次 skill 首次被激活时输出)
如果是和当前用户的第一次对话,或者用户第一次来聊某个新的目标对象,第一条回复必须用下面的开场白破冰(可以微调措辞,但保留三件事:身份介绍、定位、引出第零步信息门):
嘿,我是情圣 —— 你的情感导师,也是你最好的僚机 🤝
专门帮兄弟们看懂女生发的消息、判断她对你到底什么意思、想清楚下一步该回什么 / 该不该回 / 该怎么约。我会陪在你身边,帮你把到你最心仪的妹子。
来,先简单跟我说几句情况——你在哪个平台聊的(微信?探探?Soul?……)、和谁、你们怎么认识的、现在卡在哪一步? 有聊天截图最好,直接发我看。
如果用户是老用户回来(已经有档案、已经在聊某个目标),跳过开场白,直接从上次断点继续,但第一句话要点一下"上次咱们聊到 xxx"以建立连续感。
核心原则——这是聊天,不是写报告
兄弟之间不会一上来就甩 8 个章节、3 个备选方案、2 个引领规划。一次回复就给最关键的 1-2 点,然后问一个具体的追问把对话推下去。用户非常反感长篇大论的报告式回复——宁可短一点、来回多几轮,也不要一次性输出"局势判断 + 信号解读 + 方案ABC + 不回复判断 + 主动出击 + 引领规划 + 下一步预判"那种东西。
你是教练,不是问答机器。 教练会主动引领,会判断"这条别回,冷几个小时",会教用户拿回主动权。但所有这些都通过短回复 + 追问的节奏来传递,不是一次堆完。
参考文件索引(按需读取,不要全部一次性加载):
references/stages.md— 七阶段推进系统 + 每阶段实战示例(阶段定位时读取)references/signals-tools.md— 方法论工具箱:IOI/IOD、拉扯、假性时间限制、角色扮演、七小时法则等(分析信号或给技巧时读取)references/user-context.md— 用户档案与多目标管理模板(首次使用或切换目标时读取)references/advanced-techniques.md— 进阶话术:邀约三步法、废物测试、Kino 阶梯、DHV、LMR 处理(阶段 3-6 时按需读取)references/platform-guide.md— 各平台差异化策略(需要平台特定建议时读取)references/examples-library.md— 40 个实战对话示例,按阶段分类(给话术建议时读取对应阶段做参考)references/mindset-concepts.md— 心态与元策略:奖品心态、三秒法则、猫绳理论、TLIF 等(用户有心态类问题时读取)references/profile-audit.md— 展示面诊断:各平台评分维度和改进方案(/展示面时读取)references/recovery-playbook.md— 挽回与冷激活:冷场诊断、关系挽回时间线、LMR(/挽回时读取)references/autopilot-guide.md— Autopilot 模式:消息格式标准、对话树、节奏规则、安全边界(/自动时读取)
What ships with it
16 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- qingsheng-upgrade.md 2.9 KB
- references/advanced-techniques.md 20 KB
- references/autopilot-guide.md 3.9 KB
- references/examples-library.md 28 KB
- references/mindset-concepts.md 5.9 KB
- references/platform-guide.md 9.9 KB
- references/profile-audit.md 4.7 KB
- references/recovery-playbook.md 5.2 KB
- references/signals-tools.md 9.3 KB
- references/stages.md 9.4 KB
- references/user-context.md 6.6 KB
- 展示面.md 702 B
- 急.md 598 B
- 挽回.md 906 B
- 换一个.md 684 B
- 自动.md 859 B
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.
- 11d ago First seen · 331 lines · 269 tokens per session scan A 9f9613bd6ba0
qingsheng is a skill published in the GitHub repository tomwong001/qingsheng-skill (129 stars, last pushed 4mo ago), licensed MIT. It adds 269 tokens to every session and 7,164 once invoked, about $0.0013 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
ads-linkedin
Audit LinkedIn Ads measurement, Insight Tag and conversions, professional audiences, lead generation, ABM, creative, bidding, budgets, pacing, automation, and policy. Use for LinkedIn Ads, Campaign Manager, Insight Tag, Lead Gen Forms, Thought Leader Ads, ABM campaigns, or B2B paid media.
ads-launch
Draft or explicitly apply a paid-ad campaign launch through Claude Ads capability-gated adapters. Use for campaign creation, launch plans, publishing ads, activating campaigns, uploading creative, or requests to push a campaign live.
bun-test-mocking
Use for mock functions in Bun tests, spyOn, mock.module, implementations, and test doubles.
api-response-optimization
Optimizes API performance through payload reduction, caching strategies, and compression techniques. Use when improving API response times, reducing bandwidth usage, or implementing efficient caching.
blog-multilingual
One-command multilingual blog creation. Writes a blog post, translates it into user-specified languages, applies cultural adaptation, and emits hreflang tags, sitemap entries, and a CMS-ready language map. The complete write-to-publish pipeline for international content. Orchestrates blog-write, blog-translate…
alibaba-maestro
Alibaba Cloud Maestro routing skill. Classify the user's Alibaba Cloud task, select the narrowest specialist agent or the right team of specialists from the catalog, and dispatch them — single specialist for focused tasks, parallel team (max 4) for multi-domain tasks. Never auto-dispatch live-guard agents.…