ai-demo-builder
Skill1 starsChanged 3 months agoMITView on GitHub
Build working AI-automation demos tailored to a prospect's business to close automation sales — voice receptionist, review-request flow, lead follow-up bot, appointment reminders. Triggers - "build a demo for [prospect]", "AI demo", "automation demo", "voice receptionist demo", "review request demo". NOT for website builds — use website-builder.
In this repository47 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- access-transfer-trackerSkill
- ai-3d-generatorSkill
- ai-asset-generatorSkill
- apple-grade-designSkill
- brand-guide-builderSkill
- client-onboardingSkill
- cold-email-campaignSkill
- cold-emailSkill
- company-teardownSkill
- competitive-ads-extractorSkill
- content-calendarSkill
- content-repurposerSkill
- content-research-writerSkill
- content-strategySkill
- daily-briefSkill
- discovery-call-prepSkill
- email-composerSkill
- follow-up-nurtureSkill
- gbp-local-seoSkill
- hormozi-100m-leadsSkill
- interactive-website-builderSkill
- invoice-trackerSkill
- lead-flow-verifierSkill
- lead-scraperSkill
- mcp-integrationSkill
- meeting-listenerSkill
- post-purchase-email-seriesSkill
- productization-journalSkill
- proposal-generatorSkill
- prospect-workupSkill
- regulated-copy-complianceSkill
- retainer-reportingSkill
- review-managerSkill
- scroll-demo-builderSkill
- seo-auditSkill
- site-launch-cutoverSkill
- site-qa-checklistSkill
- social-media-auditorSkill
- social-media-managerSkill
- social-video-producerSkill
- stylized-3d-worldbuildingSkill
- video-analyzerSkill
- web-researchSkill
- website-builderSkill
- weekly-client-checkinSkill
- welcome-email-seriesSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Don't explain what AI can do — show them.
- Distinctive: line 34
Call-event webhook → serverless function: create the calendar event, send a confirmation, log the call summary to the workspace + <TASK …
- Distinctive: line 36
Test 3 scenarios (hours question, booking, services list), screen-record, then share the number: "Call this."
- Distinctive: line 85
Write-back: after any prospect demo or conversation, log an activity note on the prospect's workspace record and update the matching <TASK …
- Very specific: line 90
If a demo asset is client- or public-facing and touches a regulated topic (financing offers, health claims, legal outcomes, insurance), run …
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
AI Automation Demo Builder
Don't explain what AI can do — show them. Build a working demo in the prospect's own business context, let them interact with it live, then close. A demo they can call, text, or submit to beats any slide deck.
When NOT to use
- Website builds →
website-builder - Cold outreach copy →
cold-email
Stack (map each need to YOUR live tools before building)
| Need | Use |
| Scheduled or chat-triggered automation | Your always-on agent box (if you run one), or a scheduled cloud agent |
| Webhook endpoint / form handling / API proxy | Serverless function (keep the source dir OUTSIDE the publish dir; origin-check + rate-limit anything public) |
| SMS | Your SMS provider (e.g. Twilio) |
| Email sequence | Your email platform, or direct transactional send |
| AI generation inside a function | Your LLM router — validate model slugs against its live API first |
| Voice agent | A voice-AI platform (e.g. VAPI) + a provisioned phone number |
Demo types
1. AI Voice Receptionist
The prospect calls a number and talks to an AI that knows their business and books appointments.
- Configure the voice agent with the business's name/hours/services/FAQ — scrape their site or ask the founder for the facts; never invent capabilities
- Call-event webhook → serverless function: create the calendar event, send a
confirmation, log the call summary to the workspace +
<TASK SYSTEM> - Test 3 scenarios (hours question, booking, services list), screen-record, then share the number: "Call this."
- Dependencies: voice-platform account + phone number. Not buildable until those exist — say so upfront.
2. Automated Review Request Sequence
Job marked complete → the customer automatically gets a perfectly-timed review ask.
- Trigger: a message to your automation box or a webhook when a job completes
- Sequence: immediate thank-you text → 2h review request with the direct Google review link → 3-day gentle follow-up if no review yet
- SMS via your provider, email variant via your email platform; track asked/reviewed/response-rate in the workspace
- Demo: walk the sequence using their real business name, city, and actual review link
3. Lead Follow-Up Bot
Form submission → instant personalized AI response + owner notification within 60 seconds.
- Form → serverless function → LLM generates the personalized reply → email out
- Owner notification (chat/text/email): lead details, an urgency-signal lead score, suggested talking points
- Demo: a simple form page, submit a test inquiry live in front of them
4. Appointment Reminder System
Automated 24h (email + text) and 2h (text) reminders with confirm/reschedule replies; cancellations ping the owner immediately.
- A scheduled job reads the calendar (calendar API or a manual list) and fires the sends
- Dependencies: calendar access + SMS provider
Picking the demo
Choose by the prospect's loudest pain — missed calls / no reviews / slow follow-up / no-shows. Rough vertical defaults (primary + secondary): dental-medical → 4+1 · trades/contractors → 3+2 · restaurants → 1+2 · legal → 3+1 · real estate → 3+4 · auto repair → 2+4. Their actual complaint overrides the table.
Delivery process
- Customize with their real data — business name, hours, actual review link, industry language. This is what separates a demo from a pitch deck. Never fabricate reviews, stats, or capabilities the system doesn't have.
- Build into
ventures/<automation-venture>/deliverables/demos/<prospect-slug>/<demo-type>/(mkdir -p). Reusable vertical templates accumulate indeliverables/demos/_templates/— the second demo in any vertical should take 30 minutes, not 3 hours. - Deploy (if the demo needs hosting): get the founder's explicit green-light first,
then deploy with your pinned site ID (
netlify deploy --prod --site "<YOUR-SITE-ID>"or your host's equivalent). If a serverless function means the demo only works deployed, tell the founder BEFORE they try it locally and hit the error. - Present live (let them call the number / submit the form) or screen-recorded with narration.
- Close: "That's this running 24/7 for [Business]. Setup takes about 1–2 weeks."
Pricing is value-based and set by the founder per deal — never quote a number.
→ run
proposal-generatorfor the formal proposal. - Write-back: after any prospect demo or conversation, log an activity note on the
prospect's workspace record and update the matching
<TASK SYSTEM>task/stage before closing out.
Compliance gate
If a demo asset is client- or public-facing and touches a regulated topic (financing
offers, health claims, legal outcomes, insurance), run regulated-copy-compliance
against it before anyone outside the team sees it.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

