Unwrapped

Teardown · captions

CAPTIONS

CAPTIONS

CategoryAI Video CreatorLast round · $60M · 2024Site ↗
  • Index Ventures
  • Sequoia Capital
  • Kleiner Perkins
  • Andreessen Horowitz
UX wrapper

User footage + foundation video/voice models + mobile-first creator app.

01

Public data / API layer

User video uploads
User video uploadsYours
User audio recordings
User audio recordingsYours
Web-scraped reference footage
Web-scraped reference footageScraped
Licensed stock media
Licensed stock mediaLicensed

Internal replication score

Easy
0.73

Feasibility of a useful internal substitute built with Claude (or similar), the same data access, and light agent logic — not rebuilding the whole product.

IRS = 0.30·D + 0.25·L + 0.20·O + 0.15·R + 0.10·Sthis record · 73%
  • D

    Data accessibility

    weight 0.300.85
    • 1.0mostly customer-owned / public / standard third-party sources
    • 0.5mixed accessibility
    • 0.0hard-to-access or proprietary source layer
  • L

    LLM substitutability

    weight 0.250.70
    • 1.0mostly retrieve / prompt / cite / summarize / classify / compare
    • 0.5mixed standard + custom behavior
    • 0.0strongly custom model behavior (fine-tunes on proprietary data, etc.)
  • O

    Output simplicity

    weight 0.200.75
    • 1.0straightforward internal work product (memo, list, reply, SQL query)
    • 0.5moderately specialized
    • 0.0highly specialized (e.g. FDA-graded clinical text)
  • R

    Review / risk tolerance

    weight 0.150.80
    • 1.0internal use with human review is acceptable
    • 0.5moderate risk
    • 0.0very low tolerance for error (e.g. external legal filings)
  • S

    Surface complexity

    weight 0.10inverse — higher means less surface dependence0.30
    • 1.0a simple internal shell is enough
    • 0.5polished workflow matters somewhat
    • 0.0product surface / rollout / trust posture is central to value
LabelsEasy ≥ 0.67Medium ≥ 0.34Hard < 0.34

Missing factor rows use heuristics from wrapper scores. Editorial heuristic, not investment advice.

Build it yourself

Recreate the workflow inside your org.

Internal build

Build it yourself

Same user footage + frontier video/voice APIs + thin prompt templates — lacks mobile UX and one-tap curation.

Internal use only. Replacing them in-market is a different bar than replaying the useful workflow inside your org.

01 · Connectors & flow

User video uploads
User video uploads
User audio recordings
User audio recordings
Web-scraped reference footage
Web-scraped reference footage
Licensed stock media
Licensed stock media

Internal build map

Data in

Connectors
Connectors

Agent layer

Planner
Tools + retrieval
Reasoning model

Logic

LLM API
video generation
voice synthesis
lip-sync
caption generation
edit automation
not custom weights

Outputs

Internal search
Answer
Citations

02 · Claude / agent prompt

Paste as the system or developer message in Claude (or your agent runtime). Scroll to read; Copy grabs the full text.

Claude / agent prompt

// Video editing assistant for internal short-form content You are a video editing assistant for [YOUR_TEAM]'s social media and marketing content. You help team members transform raw footage into polished short-form videos using ONLY materials they provide: uploaded video files, audio recordings, script text, and reference images. ## What you must do 1. **Generate edits from prompts**: Accept natural language requests like "add captions with this style" or "generate B-roll for this scene" and produce structured editing instructions (trim points, overlays, caption formatting). 2. **Caption generation**: Transcribe audio, format captions with timestamp alignment, suggest styling (font, animation, color) based on brand guidelines. 3. **Asset recommendations**: When user requests B-roll, music, or effects, suggest accessible alternatives (licensed stock libraries, Creative Commons sources, or user-owned assets). 4. **Export guidance**: Recommend format, resolution, aspect ratio based on target platform (TikTok, Instagram Reels, YouTube Shorts). ## What you are not Not a replacement for human creative judgment — all edits require team review before publishing. Internal use only. ## Refusal Refuse requests to generate synthetic actors or deepfakes without explicit user consent and legal clearance. Refuse requests for copyrighted audio or footage substitution without licensing confirmation. ## Safety All video generation is for internal review only. Human approval required before external distribution.

03 · Result

Add animated captions to this 30-second clip with our brand colors
user-video

Transcribed audio, formatted captions with brand hex colors, suggested keyframe animations.