/cs:webinar — Webinar & virtual-event marketing workflow. Plan a webinar from scratch (sized backward from the business goal), rescue one whose numbers disappoi
复制下面这句话,粘贴给 Claude Code、Codex、Cursor 等 AI 编程工具,它会读取安装说明并在你确认后完成安装。
请阅读 https://ai.atlankj.com/install/asset/gh-claude-skills-1c65e0b5d70f ,按照其中的说明把「cs-webinar」安装到你(当前 AI 工具)中。执行前先告诉我将运行的命令和写入的位置,等我确认。
查看 AI 将读取的安装说明正在读取 GitHub 原文…
内容来自 GitHub 原始文件,由原作者维护。在 GitHub 查看
Command: /cs:webinar [mode] [args]
The cs-webinar command is the entry point for webinar workflows: plan → promote → run → follow up, or diagnose → fix → re-run.
/cs:launch / launch-strategyemails skillplan — Design the whole motion from scratch/cs:webinar plan
Walks the intake, locks the promise + format, sizes the funnel backward from the business goal,
builds the promotion runway, and designs show-up + live-to-close + follow-up. Delivers a full plan
using marketing-skill/skills/webinar-marketing/templates/webinar-plan-template.md.
rescue — Diagnose and fix an underperforming webinar/cs:webinar rescue --input funnel.json
Scores the funnel, names the weakest stage, and returns ranked fixes targeting the actual bottleneck — not a reflexive landing-page rewrite.
evergreen — Convert a past webinar to on-demand/cs:webinar evergreen
Maps the on-demand registration → watch → follow-up automation, with honest live-vs-simulated framing.
score — Run the funnel scorer directly/cs:webinar score --input funnel.json
/cs:webinar score # embedded sample data
| Q | Asks | When |
|---|---|---|
| Q1 | Which mode — plan / rescue / evergreen? | Always |
| Q2 | Business goal + conversion action (leads, pipeline, adoption, retention, brand)? | Always (drives the backward funnel math) |
| Q3 | Audience temperature (customers / warm / owned_cold / paid_cold)? | Always (selects benchmarks) |
Read marketing-context.md first if it exists — it covers brand voice, personas, and customer language,
so you only ask for what's specific to this event.
# Mode: rescue / score — find the broken stage first
python3 marketing-skill/skills/webinar-marketing/scripts/webinar_funnel_scorer.py funnel.json
# → overall 0-100 score + per-stage rate vs. benchmark + named bottleneck
# Pipe JSON via stdin
cat funnel.json | python3 marketing-skill/skills/webinar-marketing/scripts/webinar_funnel_scorer.py -
# Demo on embedded sample data (no --help flag — run with no args)
python3 marketing-skill/skills/webinar-marketing/scripts/webinar_funnel_scorer.py
Input JSON (registrations + attended_live required; rest optional):
{
"invited": 5000, "page_visits": 1800, "registrations": 620,
"attended_live": 180, "cta_clicks": 40, "conversions": 14,
"audience": "owned_cold", "runtime_min": 45, "avg_watch_min": 26
}
Always size from the business goal backward — this stops anyone celebrating 800 registrations while 6 people buy:
Business goal: 20 sales-qualified opportunities
÷ attendee→SQO rate (~10%) → need 200 engaged attendees
÷ register→attend (~35% live) → need ~570 registrations
÷ landing-page CVR (~40%) → need ~1,425 landing-page visits
→ promotion must drive ~1,425 qualified visits
If the required visits exceed the reachable audience, fix the goal, format, or promotion budget now.
The scorer calibrates per audience temperature (warmer audiences convert better at every stage):
| Audience | Page→Reg | Reg→Attend | Attend→CTA | Attend→Convert |
|---|---|---|---|---|
customers | 40% | 50% | 25% | 12% |
warm | 35% | 42% | 22% | 10% |
owned_cold | 25% | 35% | 18% | 7% |
paid_cold | 18% | 28% | 15% | 5% |
cs-webinar-marketerwebinar-marketing/cs:aeo (get supporting content cited by AI search), launch-strategy (full launches)Version: 2.9.0 License: MIT