// Generate 2-minute product demo videos for technical projects (GitHub repos, APIs, technical products). Runs parallel build options with structured critic review. Use when demoing code, explaining architecture, or creating explainer videos.
Local demo: City-side ADU plan review. Point this at a plan binder (PDF or pre-extracted PNGs) and a city name. It reviews the plans sheet-by-sheet against state and city code, then generates a draft corrections letter. Fire-and-forget — no human-in-the-loop pause. Triggers on: 'Review this ADU plan set for [City]' or 'Run the city review on [path]'.
Local demo: Contractor corrections analysis and response — two-phase flow with a human-in-the-loop pause. Phase 1 reads the corrections letter, researches codes, views plan sheets, and generates questions for the contractor. The agent STOPS and presents questions to the user. After the user provides answers, Phase 2 generates the full response package (response letter, professional scope, corrections report, sheet annotations). Triggers on: 'Analyze these corrections' or 'Run the contractor flow on [path]'.
City-side ADU plan review — the flip side of adu-corrections-flow. Takes a plan binder PDF + city name, reviews each sheet against code-grounded checklists, checks state and city compliance, and generates a draft corrections letter with confidence flags and reviewer blanks. Coordinates three sub-skills (california-adu for state law, adu-city-research OR a dedicated city skill for city rules, adu-targeted-page-viewer for plan extraction). Triggers when a city plan checker uploads a plan binder for AI-assisted review.
Operations manual for the CrossBeam ADU Permit Assistant. Teaches AI agents how to operate the deployed system — trigger flows, check status, read results, navigate the UI, and query the database.
Analyzes ADU permit corrections letters — the first half of the corrections pipeline. Reads the corrections letter, builds a sheet manifest from the plan binder, researches state and city codes, views referenced plan sheets, categorizes each correction item, and generates informed contractor questions. This skill should be used when a contractor receives a city corrections letter for an ADU permit. It coordinates three sub-skills (california-adu for state law, adu-city-research for city rules, adu-targeted-page-viewer for plan sheet navigation) to produce research artifacts and a UI-ready questions JSON. Does NOT generate the final response package — that is handled by adu-corrections-complete after the contractor answers questions. Triggers when a corrections letter PDF/PNG is provided along with the plan binder PDF.
City-side ADU plan review — the flip side of adu-corrections-flow. Takes a plan binder PDF + city name, reviews each sheet against code-grounded checklists, checks state and city compliance, and generates a draft corrections letter with confidence flags and reviewer blanks. Coordinates three sub-skills (california-adu for state law, adu-city-research OR a dedicated city skill for city rules, adu-targeted-page-viewer for plan extraction). Triggers when a city plan checker uploads a plan binder for AI-assisted review.
| name | product-demo-video |
| description | Generate 2-minute product demo videos for technical projects (GitHub repos, APIs, technical products). Runs parallel build options with structured critic review. Use when demoing code, explaining architecture, or creating explainer videos. |
| metadata | {"clawdbot":{"emoji":"🎬"}} |
Generate 2-minute product demo videos for technical projects with parallel build options and structured critic review.
DISCOVERY → SCRIPT → PARALLEL BUILD → PARALLEL CRITICS → ITERATE → FINAL → DELIVER
│ │ │ │ │ │ │
5-10m 10-15m 30-60m 10-15m 0-3 loops gate ship it
Clone and analyze the repo:
gh repo clone <owner/repo> /tmp/<repo>
Extract:
Ask (or use defaults):
| Question | Default |
|---|---|
| Duration | 2 min |
| Audience | Mixed (tech + business) |
| Tone | Professional-conversational |
| Format | Animated explainer |
Use references/script-template.md structure:
Scene 1: Hook (0:00-0:15) — Problem statement
Scene 2: Context (0:15-0:30) — Why this matters
Scene 3-5: Features (0:30-1:30) — 3 key capabilities, 20s each
Scene 6: Technical (1:30-1:50) — Architecture/credibility
Scene 7: CTA (1:50-2:00) — What to do next
Guidelines:
Spawn 2 sub-agents to build options simultaneously:
// Option A: Remotion (animated)
sessions_spawn({
label: "demo-video-option-a",
task: `Build animated demo video using Remotion.
Script: [paste script]
Voice: Use ElevenLabs, voice ID CcEB0ARKH8qIyzh2I1OR
Output: /tmp/demo-video/option-a.mp4
Approach:
1. Generate voiceover per scene (ElevenLabs API)
2. Create Remotion project with scene components
3. Sync animations to audio timing
4. Render at 1080p 30fps`
})
// Option B: Slides + ffmpeg (simple)
sessions_spawn({
label: "demo-video-option-b",
task: `Build slide-based demo video.
Script: [paste script]
Voice: Use ElevenLabs, voice ID CcEB0ARKH8qIyzh2I1OR
Output: /tmp/demo-video/option-b.mp4
Approach:
1. Generate voiceover (ElevenLabs API)
2. Create slides with Puppeteer (HTML → PNG)
3. Combine with ffmpeg, sync to audio
4. Add simple transitions`
})
Wait for both. Present options to user or auto-select based on quality.
Spawn 2 critic personas (see references/ for full rubrics):
// Business Critic
sessions_spawn({
label: "critic-business",
task: `Review demo video as Business Critic "Mike".
Video: [path]
Script: [content]
See references/critic-business.md for full persona and rubric.
Score: Clarity, Professionalism, Business Value (1-10 each)
Output: Structured review with overall score and specific fixes.`
})
// Technical Critic
sessions_spawn({
label: "critic-technical",
task: `Review demo video as Technical Critic "Senior Engineer".
Video: [path]
Script: [content]
See references/critic-technical.md for full persona and rubric.
Score: Technical Accuracy, Credibility, Completeness (1-10 each)
Output: Structured review with overall score and specific fixes.`
})
Quality gates:
| Gate | Action |
|---|---|
| Both critics ≥7.5, no Critical issues | → Ship |
| Any critic <7.0 OR Critical issue | → Iterate |
| After 3 iterations, still failing | → Escalate to human |
Iteration loop:
Before shipping, verify:
Output:
✅ Demo video complete!
📹 Video: demo-video-final.mp4 (2:03)
🎯 Audience: [target]
📊 Scores: Business 8.2/10 | Technical 7.8/10
Covers:
1. [Key point 1]
2. [Key point 2]
3. [Key point 3]
Files saved to: [location]
Deliver to:
| Tool | Purpose | Required |
|---|---|---|
| ElevenLabs | Voiceover generation | Yes |
| Remotion | React video framework (Option A) | One of A/B |
| ffmpeg | Video processing (Option B) | One of A/B |
| Puppeteer | Slide screenshots (Option B) | One of A/B |
| sessions_spawn | Parallel sub-agents | Yes |
Voice default: CcEB0ARKH8qIyzh2I1OR (Jeremy - Technical Tutorial Narrator)
references/critic-business.md — Business critic persona & rubricreferences/critic-technical.md — Technical critic persona & rubricreferences/script-template.md — Scene-by-scene script structuretemplates/remotion/ — Remotion project scaffold with componentsscripts/generate-voiceover.py — ElevenLabs TTS helperscripts/render-slides.sh — ffmpeg slide-to-video combiner