Use when 王建硕 wants to systematically grow his X (Twitter) followers by running numbered, A/B-testable growth experiments and tracking which ones actually work. Every action gets a number, a hypothesis, a target metric, a before-state (for rollback), and a verdict. The North-Star metric is the new-follower ÷ profile-visit ratio (conversion) — every profile change is judged against it. Daily it ingests the X Analytics CSV export, scores each running experiment, and recommends keep / rollback. Triggers — "涨粉", "增加 X 粉丝", "X 涨粉实验", "follower growth", "A/B test my profile", "今天的涨粉检查", "/wjs-x-increasing-follower".
npx skills add https://github.com/jianshuo/claude-skills --skill wjs-x-increasing-follower
把「涨粉」当工程做:每个改动是一个带编号的实验,有假设、有目标指标、有 before(可回滚)、有判决。不靠感觉,靠数。
单账号没法做平行 A/B —— 只能做时间轴上的前后对比。 所以唯一可信的北极星指标是 转化率 = 新增关注 ÷ 主页访问(ratio):它对爆款流量免疫。一条推爆了带来一堆访问,ratio 不一定动;但 bio 改好了,每个来访的人更愿意关注,ratio 一定动。
所以指标分层(每个 action 必须声明自己被哪个指标考核):
| Action 类型 | 拨动的杠杆 | 用什么考核 |
|---|---|---|
| profile(bio / 名字 / 头像 / banner / 置顶 / URL / 地点) | 转化 | ratio(北极星,抗爆款) |
| posting(格式 / 钩子 / 频率 / thread vs 单条) | 触达 | profile visits + impressions(ratio 当护栏,别把转化拖垮) |
| engagement(回复 / 关注别人 / 互动) | 触达 | new follows + visits |
| timing(发布时间) | 触达 | profile visits |
诚实护栏(写死在 evaluate.py 里): 用中位数不用均值(一天爆款骗不了判决);够 7 天 / 够数据才下判决;同指标实验窗口重叠 → 打 confounded 标记;只给「方向性」结论,不号称因果。
回滚是一等公民。 每个 action 存了精确的 before 值,#N 永远能还原。判 ❌ rollback 先问王建硕,绝不静默改他的 bio。
ratio 这个数 X API / xurl 拿不到 —— 只活在 Analytics 看板里。所以靠 CSV 导出:
打开 x.com/i/account_analytics → Overview → 右上角下载图标(7D/2W/4W/3M/1Y 想要哪段先选好)→ 导出 CSV → 丢进 inbox/(或直接给路径)。
/wjs-x-increasing-follower/schedule daily /wjs-x-increasing-follower(见末尾「每日检查」)/wjs-tweeting-from-articles 或直接 xurl post/wjs-promoting-skills / /publish-skill/wjs-syndicating-articles所有脚本在 scripts/,状态在 state/。先 cd 到 skill 目录。
python3 scripts/ingest-csv.py /path/to/export.csv
脚本模糊匹配列名(Profile visits / New follows / Impressions / Followers),按日期 upsert,自动算 ratio。它会打印它认的列映射 —— 看一眼对不对,不对就 --visits-col "X" --follows-col "Y"。重跑更长的导出会覆盖重叠的天,安全。
看 state/daily.jsonl 现状,给王建硕一份带编号的实验菜单。每条必须有:category / title / hypothesis / metric。原则:
入账(自动分配编号,别手改 jsonl):
python3 scripts/ledger.py add --category profile --metric ratio \
--title "Bio: 开头一行放硬证明" \
--hypothesis "访客看到具体成绩比看到口号更愿意关注"
实验菜单参考(按杠杆分):
state/SCOREBOARD.md 的「To-do」区就是这份清单的落地版(见 Step 6)。
先抓 before-state(profile 字段):
xurl "/2/users/me?user.fields=name,description,url,location,profile_image_url,pinned_tweet_id"
记下当前值。然后真正去改:
xurl --auth oauth1 -X POST "/1.1/account/update_profile.json?description=<urlencoded>"
(name / url / location 同理。失败/没 enroll OAuth1 → 退回 App 里手改,照样记账。)
记账(盖上 applied 日期 + before 值):
python3 scripts/ledger.py apply 1 \
--before-field description --before-value "<原 bio 原文>" \
--after-value "<新 bio>"
之后这个实验进入测量期:baseline = applied 前 window_days(默认7) 天的指标中位数;判决要等 applied 后攒够数据。
scripts/daily-check.sh
它会:ingest inbox/ 里的新 CSV(用完归档到 inbox/done/)→ evaluate.py --write-verdict 把方向性判决写回账本 → 重生成 SCOREBOARD.md。注意:它只写判决,不动状态 —— keep/rollback 是人的决定。
python3 scripts/evaluate.py # 看每个在跑实验:baseline vs post,Δ%,verdict
判决(相对 baseline 中位数):Δ ≥ +10% → keep;Δ ≤ -10% → rollback;之间 → flat。注意 ⚠ confounded / ⚠ thin baseline 标记,有就别太当真,延长窗口或停掉重叠实验再测。
python3 scripts/ledger.py keep 1 --note "+58% 转化"apply 时存的 before 值还原(API 或手改),再python3 scripts/ledger.py rollback 1 --note "转化掉了,还原 bio"(命令会打印要还原的 before 值)
python3 scripts/scoreboard.py # 写并打印 state/SCOREBOARD.md
四块:📈 现状(最新 ratio / 7 日中位 ratio / 粉丝数)、🧪 在跑实验(baseline/post/Δ/判决/告警)、✅ To-do 待办(proposed 清单)、📚 已结案(kept / rolled_back)。给王建硕看就发这个文件。
daily.jsonl —— 一天一行:{date, profile_visits, new_follows, ratio, impressions, followers_total}。actions.jsonl —— 一实验一行:{id, category, title, hypothesis, metric, window_days, before:{field,value}, after:{value}, applied, status, evaluated, verdict, notes}。状态流:proposed → active → (kept | rolled_back)。SCOREBOARD.md —— 生成物,人看的。--window-days 默认 7 天--threshold 默认 ±0.10(相对 ±10%)--min-post-days 默认 3 天/schedule daily /wjs-x-increasing-follower —— 每天调起本 skill:跑 daily-check.sh,把 SCOREBOARD.md 现状 + 任何判 ❌ rollback 的实验 发给王建硕并征求是否回滚。CSV 还是得他手动导出丢进 inbox/(ratio 数据 API 拿不到,这步绕不开)。
NGS analysis toolkit. BAM to bigWig conversion, QC (correlation, PCA, fingerprints), heatmaps/profiles (TSS, peaks), for ChIP-seq, RNA-seq, ATAC-seq visualization.
Automate customer engagement workflows including broadcast triggers, message analytics, segment management, and newsletter tracking through Customer.io via Composio
NGS analysis toolkit. BAM to bigWig conversion, QC (correlation, PCA, fingerprints), heatmaps/profiles (TSS, peaks), for ChIP-seq, RNA-seq, ATAC-seq visualization.
Query live pathogen genomic surveillance data through the GenSpectrum LAPIS API to find which viral lineages are circulating now, how fast they are growing, and what mutations they carry. Use whenever a question depends on the current state of a pathogen population rather than on remembered facts - which SARS-CoV-2 variant is dominant, whether a Pango lineage is still designated or has been withdrawn, what clade or genotype of H5N1 is in a host or region, whether a PCR primer or assay target still matches circulating sequence, or how a lineage's prevalence has moved week to week. Triggers include "variant surveillance", "genomic surveillance", "what variant is circulating", "dominant variant", "Pango lineage", "lineage prevalence", "growth advantage", "SARS-CoV-2 variant", "XFG", "clade 2.3.4.4b", "H5N1 genotype", "influenza clade", "RSV/mpox/measles/dengue lineage", "CoV-Spectrum", "LAPIS", "Nextclade", "pango-designation", and any request to report what a pathogen population looks like today.
NGS analysis toolkit. BAM to bigWig conversion, QC (correlation, PCA, fingerprints), heatmaps/profiles (TSS, peaks), for ChIP-seq, RNA-seq, ATAC-seq visualization.
Create a comprehensive product strategy using the 9-section Product Strategy Canvas — vision, segments, costs, value propositions, trade-offs, metrics, growth, capabilities, and defensibility. Use when building a product strategy, creating a strategic plan, or defining product direction.
Build a marketing performance report with key metrics, trend analysis, wins and misses, and prioritized optimization recommendations. Use when wrapping a campaign, when preparing weekly, monthly, or quarterly channel summaries for stakeholders, or when you need data translated into an executive summary with next-period priorities.
Calculate SaaS revenue, retention, and growth metrics. Use when diagnosing momentum, churn, expansion, or product-market-fit signals.
Take jianshuo/wjs-x-increasing-follower from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.