Pull Quote Beside the Speaker — effect skill for quotes & reviews
For AI agents: find the right skill in one call —
curl -s "https://varg.ai/how-to/api/search?q=<what the person wants>"returns the matching how-to page and the best skills with install commands, as markdown (Russian works too). Full index: https://varg.ai/how-to/llms.txt. Every page has a markdown twin: add.mdto its path. Before generating anything, show the person the options on the how-to page.
While a customer talks on camera, their key sentence appears as large type in the empty side of the frame.
- Type: Effect · Quotes & reviews
- Slug:
pull-quote-beside-speaker - Status: live
- Publisher: HeyGen (community)
- License: Apache-2.0
- Source: https://github.com/heygen-com/hyperframes/tree/main/registry/blocks/lt-dark-card
- Works in: Claude Code, Claude.ai, Codex, Cursor, OpenClaw, Gemini CLI, Any MCP client, varg Studio
- Requires: ffmpeg, HyperFrames (npx hyperframes), Remotion, CapCut (by hand)
- Cost: Free — the take renders locally; generated footage, if any, bills by the model
What it does
Built on Lower Third — Dark Card by HeyGen (HyperFrames), Apache-2.0. The write-up and recipes below are varg's.
Speaker framed left or right third; the strongest line fades up per word on the open side, high contrast, no card needed.
When: Video testimonials and interviews, so the claim survives sound-off viewing.
Watch out: Do not cover the speaker's face or hands, and do not repeat every word they say — only the pull quote.
Timing: Start with the spoken words; per-word fade 0.2–0.3 s; hold to the end of the sentence + 0.5 s.
Also called: testimonial overlay, quote over interview, speaker quote
In FFmpeg
-vf "drawtext=text='Cut our onboarding in half':x=w*0.55:y=h*0.4:fontsize=72:fontcolor=white:alpha='min(1,max(0,(t-8.2)/0.3))':enable='between(t,8.2,12)'"
In HyperFrames
npx hyperframes add soft-blur-in
Time each word's tl.from to its transcript timestamp; keep the text inside the open half of the frame.
In Remotion
// npx shadcn@latest add @remocn/per-word-crossfade
const o = interpolate(frame,[w.start*fps, w.start*fps+6],[0,1],{extrapolateLeft:'clamp',extrapolateRight:'clamp'});
In CapCut
Captions → Auto captions, delete all but the pull-quote line, restyle it large on the open side; Animation In "Fade in by word".
Ready-made components
- Lower Third — Dark Card (hyperframes): npx hyperframes add lt-dark-card
- Per-Word Crossfade (remocn): npx shadcn@latest add @remocn/per-word-crossfade
- Soft Blur In (animata): npx shadcn@latest add https://animata.design/r/text/soft-blur-in.json
Where this comes from
- Morphic: how to make a customer testimonial video
- CapCut: turn testimonials into quote graphics for short-form
Use when
- Video testimonials and interviews, so the claim survives sound-off viewing.
Not for
- Do not cover the speaker's face or hands, and do not repeat every word they say — only the pull quote.
Install
- Claude Code:
claude plugin marketplace add vargHQ/skills && claude plugin install varg@varg - Codex:
npx skills add vargHQ/skills --skill varg-video-edit-library -a codex -y - Cursor:
npx skills add vargHQ/skills --skill varg-video-edit-library -a cursor -y
Example output
https://s3.varg.ai/how-to/previews/d2553303cbde/take-pull-quote-beside-speaker.mp4 — The Lower Third — Dark Card component rendered as installed (npx hyperframes add lt-dark-card), defaults only (credit: Rendered by varg · HyperFrames Lower Third — Dark Card, https://hyperframes.heygen.com/catalog/blocks/lt-dark-card)
More quotes & reviews skills
- Tweet as Proof Card: A real post from a user animates in as a card: avatar, name, text, then counts tick up.
- Quote at Reading Pace: One strong customer sentence reveals word by word, then the name, role and avatar settle under it.
- Star Rating Sweep: Five stars fill left to right to the real score while the number counts up beside them.
- Wall of Love Marquee: Two or three rows of review cards drift in opposite directions, fading at the edges.
- Community Post as Proof: A real community post or comment card pops over the footage with its upvotes counting up.
- Reviews Arrive as Notifications: Short reviews land as system notifications that stack and push each other down.
Run the generation on varg
varg is one API key for the models these skills drive — Seedance 2.5, Kling 4, Veo 3.1, Wan 3, Nano Banana 2.1, GPT Image 2.5, ElevenLabs v4 and 200+ more — billed in credits, with a cost estimate before anything runs.
- Claude Code:
claude plugin marketplace add vargHQ/skills && claude plugin install varg@varg(skill + hosted MCP) - Codex, Cursor, Gemini CLI, OpenClaw:
npx skills add vargHQ/skills --skill varg-ai -a <agent> -y - Grok and any agent that reads SKILL.md:
npx skills add vargHQ/skills --skill varg-ai, or point it at the skill's SKILL.md - Claude.ai, ChatGPT, any MCP client: connect
https://mcp.varg.ai/mcp - API key: https://varg.ai/dashboard/api · Model specs: https://varg.ai/models/llms.txt