Install the "seedance-testimonial-story" agent skill from https://github.com/rediumvex/ai-video-generator-claude/tree/main/skills/08-testimonial-story into .claude/skills/seedance-testimonial-story/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "seedance-testimonial-story", then confirm the skill loads.
Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
skills CLI
$ npx skills add rediumvex/ai-video-generator-claude --skill seedance-testimonial-story -a codex
Project install goes to .agents/skills/; add -g for ~/.codex/skills/.
Install the "seedance-testimonial-story" agent skill from https://github.com/rediumvex/ai-video-generator-claude/tree/main/skills/08-testimonial-story into .agents/skills/seedance-testimonial-story/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "seedance-testimonial-story", then confirm the skill loads.
Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
skills CLI
$ npx skills add rediumvex/ai-video-generator-claude --skill seedance-testimonial-story -a cursor
Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.
Install the "seedance-testimonial-story" agent skill from https://github.com/rediumvex/ai-video-generator-claude/tree/main/skills/08-testimonial-story into .cursor/skills/seedance-testimonial-story/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "seedance-testimonial-story", then confirm the skill loads.
Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
skills CLI
$ npx skills add rediumvex/ai-video-generator-claude --skill seedance-testimonial-story -a gemini-cli
Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.
Install the "seedance-testimonial-story" agent skill from https://github.com/rediumvex/ai-video-generator-claude/tree/main/skills/08-testimonial-story into .gemini/skills/seedance-testimonial-story/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "seedance-testimonial-story", then confirm the skill loads.
Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
skills CLI
$ npx skills add rediumvex/ai-video-generator-claude --skill seedance-testimonial-story -a github-copilot
Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.
Install the "seedance-testimonial-story" agent skill from https://github.com/rediumvex/ai-video-generator-claude/tree/main/skills/08-testimonial-story into .github/skills/seedance-testimonial-story/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "seedance-testimonial-story", then confirm the skill loads.
GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
skills CLI
$ npx skills add rediumvex/ai-video-generator-claude --skill seedance-testimonial-story -a opencode
OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
Install the "seedance-testimonial-story" agent skill from https://github.com/rediumvex/ai-video-generator-claude/tree/main/skills/08-testimonial-story into .opencode/skills/seedance-testimonial-story/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "seedance-testimonial-story", then confirm the skill loads.
OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
Skill name
seedance-testimonial-story
GitHub stars
409
Token cost
~7.2k tokens
SKILL.md length
729 words
Files
1
Skills in repo
10
Repo updated
First seen
Licence
MIT
At a glance
Generate customer testimonial and social proof video prompts for Seedance 2.0 on Higgsfield.
Works in 8 steps: Lead with the result, not the person.… → Make every quote earn its screen time. A… → Specificity is the currency of… → …
Customer stories
SKILL.md covers Seedance 2.0 on Higgsfield —…, Why Traditional Testimonials…, The 2-Second Hook for… and Visual Formats for Testimonial…, plus 5 more sections
Calls black
What it does
Seedance Testimonial Story is an agent skill from rediumvex/ai-video-generator-claude. Generate customer testimonial and social proof video prompts for Seedance 2.0 on Higgsfield. Use for customer stories, case study videos, review showcases, social proof content, success story videos, or any content featuring customer results and experiences. Triggers on testimonial, review, case study, social proof, customer story, success story, results, feedback, client.
Its SKILL.md is about 7.2k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Media & Creative, covering AI video generation, Storytelling and Marketing psychology. It works with Seedance. The repository describes itself as: 10 Claude skills that generate studio-quality AI video prompts for Seedance 2.0 on Higgsfield. Viral hooks, SaaS demos, personal brand, faceless content, luxury aesthetic & more. The licence is MIT.
When your agent uses it
Customer stories
Case study videos
Review showcases
Social proof content
Example prompts
“/seedance-testimonial-story”
Workflow steps
8 steps, taken from the first numbered list in SKILL.md.
1Lead with the result, not the person. Viewers decide in the first two seconds whether to trust what follows. Give them the evidence first…
2Make every quote earn its screen time. A quote that lasts longer than 3 seconds needs to work visually as hard as it works verbally. Pair…
3Specificity is the currency of credibility. "I doubled my revenue" is weaker than "We went from $12K to $31K MRR in 11 weeks." Prompt for…
4Customer voice is the primary audio signal. If you have a voice recording, it goes first in the mix. Music and sound effects serve the…
5Match the visual register to the audience. B2B testimonials require authority signals: clean environments, professional lighting, data…
6Star ratings need a source. Five stars with no platform attribution reads as fabricated. Always include the platform name — even small…
7The final frame is a commitment device. The last image the viewer sees before deciding to act should be: product in context, or the…
8Avoid the endorsement pose. A customer facing directly into the camera with a wide smile, speaking scripted praise, is the exact visual…
What it can do on your machine
Read from SKILL.md and the folder at commit ffdad7d. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
black
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Seedance Testimonial Story loads about 7.2k tokens when it runs. Until then it costs about 101 tokens; SKILL.md has 729 words of instructions outside code blocks.
Always· name and description, kept in context so the agent knows when to use it
~101
When it runs· the whole SKILL.md, loaded when a task matches
~7.2k
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
Download SKILL.mdSave it as .claude/skills/seedance-testimonial-story/SKILL.md (or your agent's skills folder).
name
seedance-testimonial-story
description
Generate customer testimonial and social proof video prompts for Seedance 2.0 on Higgsfield. Use for customer stories, case study videos, review showcases, social proof content, success story videos, or any content featuring customer results and experiences. Triggers on testimonial, review, case study, social proof, customer story, success story, results, feedback, client.
Testimonial Story — Customer Social Proof Video Prompts
Seedance 2.0 on Higgsfield — Input Specs
Images: Up to 9 (customer photo, product screenshots, result charts, brand assets, star ratings)
Videos: Up to 3 (customer-recorded clips, screen recordings, existing brand video)
Audio: Up to 3 (customer voiceover clip, background music, ambient sound)
Output: 4-15 seconds, 720p with synchronized audio
Reference Syntax:@material[name] for uploaded assets
Why Traditional Testimonials Fail
Traditional testimonials fail because they're static talking heads in flat lighting. Cinematic social proof makes viewers feel the transformation. Show the result first, the emotion second, the person third.
The 2-Second Hook for Testimonials
The first two seconds determine whether the viewer stays. Every testimonial video needs an immediate proof signal.
Hook Patterns
Hook
Mechanism
When to Use
The Result First
Open on the metric, number, or outcome — then reveal the person behind it
Strong results with specific numbers
The Emotion Capture
Open on a genuine reaction moment — a smile, a gasp, a laugh — before context
High-emotion stories with visual customer footage
The Number Impact
Large, bold statistic fills the frame, animated to count up or slam in
Quantitative case studies with clear ROI
The Contrast Story
Split-second flash of "before" state immediately followed by "after"
Transformation stories with clear before/after
Hook Templates
THE RESULT FIRST:
"Single bold statistic center-frame. Large white typography on dark
background. Number counts up rapidly — 0 to 347% — in 1.2 seconds.
Each digit change accompanied by soft click sound. Camera slight push-in
during count. At 1.5s: number resolves. Holds for 0.3s. Then: cut to
or dissolve into customer's face — warmly lit, looking slightly off-camera,
beginning to speak. Viewer is already primed: this person got that result."
THE EMOTION CAPTURE:
"Open on tight close-up of customer's face at moment of genuine positive
reaction — mid-laugh, eyes lit, natural expression. Warm key light from
slight left angle. Shallow depth of field, background is soft-bokeh
environment related to their work or home. No speech yet — just the face.
Hold 0.8s. Then pull back slightly to include context. Sound: ambient
environment fills in, then music enters soft underneath. The emotion
happened before the words. The words explain what we already felt."
THE NUMBER IMPACT:
"Black frame. Single word appears — the relevant metric category
(REVENUE / HOURS SAVED / CLIENTS ADDED) — in thin caps, gray.
0.5s pause. Then: the number SLAMS in from below, large, white, kinetic.
Overshoot bounce physics on landing. Camera shakes 2px on impact.
Sound: deep bass thud synced to number landing. Reverb tail.
Number holds 0.8s. Then secondary text fades in below:
customer name, company, timeframe. Credibility layered after impact."
THE CONTRAST STORY:
"Split-screen. Left side: grainy, slightly desaturated footage or image
representing the 'before' state — the problem, the struggle, the mess.
Right side: clean, warm, vibrant footage representing 'after.' Both sides
hold for 1.5s. Then: left side collapses toward center, right side expands
to fill full frame. Sound: before side has flat, muted audio. Transition
whoosh. After side has rich, warm audio treatment. The visual argument
is already made before a word is spoken."
Visual Formats for Testimonial Videos
The Documentary Interview
The gold standard for testimonial credibility. Feels earned, not staged.
DOCUMENTARY INTERVIEW TEMPLATE:
"Customer seated slightly off-center — rule-of-thirds composition, looking
toward empty space in frame (not directly at camera). Interview framing:
head and shoulders visible, some environment in background.
Background: deliberately chosen to reinforce context — workspace for
B2B testimonial, home environment for consumer product, outdoors for
lifestyle brand. Background should be in soft bokeh, identifiable but
not distracting.
Camera: slightly below eye level (more authority) or eye level
(more intimacy). Static or very slow drift — no handheld jitter.
Lighting: warm interview setup. Key light 45 degrees from subject's face,
soft modifier. Fill light on shadow side, 1.5 stops under. Optional:
practical lamp or window in background for depth.
On-screen text: customer name and result appear as lower-third
approximately 2 seconds into clip. Clean sans-serif typography.
Small logo mark if applicable.
Sound: customer voice is primary. Light room tone/ambient underneath.
Optional subtle music bed — below the voice, supporting emotional register.
Voice should be crystal clear. No reverb. No noise."
The Data Visualization
For B2B and SaaS social proof. Numbers made visceral.
DATA VISUALIZATION TEMPLATE:
"Dark background, premium data environment. Chart or graph central to frame.
Data is not static — it builds, grows, animates to tell the story.
Opening: flat baseline (before state). Chart slowly builds the historical
context — flat growth, struggle plateau. Muted colors, minor key ambient.
At transition point: color shift. Chart line diverges upward sharply.
Color warms, saturates. Growth curve accelerates. Sound: ascending tone,
subtle but audible.
At destination: the current result. Number appears adjacent to endpoint
of the line or bar. Large, clear, warm-colored.
Camera slowly pushes in toward the data throughout. Customer photo or
logo appears in lower-third as attribution for the result.
Optional: voice-over of customer describing the moment of change —
synced to the transition inflection point in the chart.
Typography: minimal, clear. Result numbers are 2x the size of labels.
Timeframes labeled on axis. One visual message only — do not clutter."
The Split Journey
Before and after as parallel narratives. Emotional and visual contrast.
SPLIT JOURNEY TEMPLATE:
"Frame divided horizontally. Top half: 'before' world. Bottom half:
'after' world. Both sides run simultaneously, showing parallel actions
or states.
Top (before): slightly desaturated. Environment is cluttered or stressful.
Person's movement is hurried, tense. Color grade: cool, 6000K+.
Actions: multiple steps, visible friction. Camera: slight handheld to
convey instability.
Bottom (after): full color saturation. Environment is clean, ordered.
Person's movement is relaxed, deliberate. Color grade: warm, 3200K.
Actions: single step, effortless. Camera: locked off, stable.
Dividing line: can be a hard edge, a gradient, or a thin branded line
element. Option: the line moves through the frame like a reveal, with
the 'after' state expanding.
At resolution (3/4 through video): bottom half expands to fill full frame.
Top half disappears. The 'after' world takes over. Sound: background noise
of 'before' fades out. Warm audio environment of 'after' fills in.
Customer quote appears as text overlay."
The Montage Proof
Multiple customers or multiple results in rapid sequence. Creates overwhelming evidence.
MONTAGE PROOF TEMPLATE:
"Rapid sequence of social proof elements. 3-7 shots, 1-2 seconds each.
Each shot: different customer, same transformation narrative.
Shot structure per customer: face OR result metric. Alternate — face,
metric, face, metric. Creates visual rhythm.
Face shots: consistent lighting treatment across all — signals production
intentionality, makes diverse customers feel curated. Quick zoom-in or
push-in during each shot for energy.
Metric shots: bold typography, one number per frame, consistent type style.
Numbers are large — primary focal point.
Transitions: match-cut on similar compositions. Avoid random cuts.
Face-to-face: cut on eye-line. Metric-to-metric: cut on number position.
Sound: music track with clear BPM structure. Each cut lands on a beat.
Optional: brief audio snippet from each customer (0.5s) that blends into
overall mix — creates texture without full sentences.
End frame: compilation stat — 'Join 2,400+ customers.' or summary metric.
Warm, wide shot. Music resolves."
Typography and Text Overlay Techniques
Displaying Quotes
QUOTE OVERLAY — MINIMAL:
"Quote text appears as large white serif or sans-serif. Weight: bold or black.
Centered or left-aligned. Max 15 words per screen — cut longer quotes.
Text enters: fade-in over 0.3s, or slides up from below. Holds for
reading time (calculated: word count ÷ 3 = seconds needed). Exits: fade
or dissolve on transition. Attribution line (name, title, company) appears
below in smaller, lighter weight, 1 second after main quote settles."
QUOTE OVERLAY — DRAMATIC:
"One word or phrase at a time. Each key word appears large, centered.
Camera or background reacts to each word — subtle push or pull.
Word duration: 0.5-0.8s each. Final phrase: holds longer, larger size.
Sound: soft click or tone on each word appearance. Creates reading rhythm.
Color: white on dark, or brand color on neutral. Never reverse — dark on
light reads poorly on video."
Star Ratings
STAR RATING ANIMATION:
"Five stars arranged horizontally. Stars appear left to right, 0.15s stagger.
Each star: small scale-from-zero animation with slight overshoot. Color:
gold/amber (#F5A623 or similar). Optional: particle burst on fifth star landing.
Sound: ascending five-note arpeggio, one note per star. Final note is higher,
brighter. Stars hold for 1s then either remain as overlay or transition out.
Platform attribution (G2, Trustpilot, App Store) appears below in small text."
Metrics and Numbers
COUNT-UP NUMBER:
"Number begins at zero, counts up to target value. Speed: slow start,
accelerates, then decelerates into final number (ease-in-out curve).
Duration: 1.5-2.5s for numbers under 1,000. For large numbers: start
at ~80% of final value and count last 20% only — captures the drama
without waiting.
Typography: large, bold, sans-serif. White or brand color.
Unit (%, x, $) in slightly smaller size positioned to upper-right of number.
Context label in smaller size below: 'increase in MRR' or 'hours saved per week.'
Sound: rapid soft ticks accelerating and decelerating with the count."
Lighting for Testimonials
Interview Warm
The standard for trustworthy, human social proof. Conveys authenticity.
"Classic interview three-point lighting for testimonial video.
Key light: large soft source (simulated softbox or window) at 45-degree
angle to subject's face, slightly above eye level. Warm color temperature:
3200-4000K. This is the main illuminating source.
Fill light: on the shadow side of face. 1.5 to 2 stops dimmer than key.
Slightly cooler than key (4500K) — creates subtle color contrast between
lit and shadow sides that gives dimension.
Rim/hair light: from behind and above subject. Separates head from
background. Thin edge of warm light on shoulder and hair. Not a halo —
subtle.
Background: 1 stop darker than subject. Optional: practical light in
background (desk lamp, window glow) to give depth and environment context.
Result: subject appears warmly lit, three-dimensional, trustworthy.
Skin tones are flattering. Eyes catch the key light — the 'catchlight'
that makes a face feel alive."
Documentary Natural
Feels unplanned and authentic. Highest trust signal for skeptical audiences.
"Simulated available-light documentary treatment.
Primary source: window light from one side. Hard-ish quality — not
perfectly diffused. Creates a natural falloff across subject's face.
No fill, or very minimal fill on shadow side (1/4 reflected light only).
The asymmetry reads as real, not studio.
Color temperature: matches simulated time of day — morning (5500K, slightly
blue-white), midday (6500K, neutral), golden hour (3200K, warm orange).
Commit to one — no mixed sources.
Background: environment visible, in focus enough to be readable but not
sharp. Real spaces: kitchen table, coffee shop, workshop, outdoor setting.
Nothing generic.
Camera: very slight handheld quality — not shaky, but not locked-off.
0.5-1px micro-movement that communicates 'someone was there filming this.'
Grade: minimal processing. Slight desaturation (10%). Natural contrast.
No heavy LUT. Feels like the phone captured it, but better."
Success Glow
For triumph moments, results reveals, and high-emotion payoffs.
"Warm, elevated lighting treatment that signals victory and transformation.
Key light: golden-warm (2700-3000K), positioned slightly lower than
standard interview — 30 degrees above eye level rather than 45.
This angle catches the face differently — more like sunrise/sunset,
less like office overhead.
Background light: significantly warmer than subject light (2000-2500K),
or a practical warm glow element (candle, fireplace, warm lamp).
Background appears luminous, almost glowing.
Exposure: slightly overexposed compared to interview norm — 1/3 stop hot.
Highlights clip gently. Creates the visual language of 'bright' and
'successful.'
Optional element: subtle lens flare or light streak from key light
direction. Small — not a J.J. Abrams production. Just a hint that the
light is real and present.
Grade in post: lift the shadows slightly (not crush them). Add warmth
in midtones. Final image should feel like the best day of that person's life."
Sound Design for Testimonial Videos
Voice-First Design
In testimonial videos, the human voice is always primary. Everything else serves it.
VOICE-FIRST MIX:
"Customer voice sits at -6dB in the mix. Clear, present, centered stereo.
No reverb — voice should feel intimate and close, not roomy.
Eq: gentle high-pass at 80Hz (remove rumble), slight presence boost
at 3-5kHz (intelligibility). De-ess if needed — sibilance is fatiguing.
Music bed: -18 to -20dB under voice. Listener should barely be aware
of music consciously. It should only register when the voice pauses.
Music should never fight for the same frequency space as the voice —
avoid music with heavy midrange when voice is present.
Sound effects (if any): brief, purposeful. A notification sound for
a result metric. A soft chime on a star rating. These punctuate,
they do not dominate.
The test: can you understand every word without concentration? If yes,
the mix is right."
Ambient Storytelling
Sound environment that tells the 'after' story without words.
AMBIENT TESTIMONIAL SOUNDSCAPE:
"For B2B/work context: light keyboard sounds, distant office ambience,
occasional notification sound that resolves positively. Signals
productivity, accomplishment, flow state.
For consumer/lifestyle: warm room tone, distant city or nature ambience,
coffee sounds, comfortable environment signals. Signals ease, comfort,
elevated life.
For high-energy results: subtle energy in the room tone — a crowd reference,
outdoor air, movement and vitality. Signals a life in motion.
Build: ambient starts low (barely audible). As testimonial emotional
peak approaches, ambient subtly rises +3dB. Creates unconscious sense
of building energy. At resolution: ambient settles back, music may
swell slightly. Clean, satisfying completion."
Emotional Music Timing
MUSIC TIMING FOR TESTIMONIALS:
"Music enters under opening visual hook — soft, establishing tone.
If opening is 'before' state: music should be slightly tense or neutral.
If opening is 'after' state: music should be warm and aspirational from start.
At testimonial credibility moment (result reveal, key quote): music adds
an element. A second instrument enters, or the arrangement becomes slightly
fuller. The harmonic environment enriches — unconscious signal of elevation.
At final frame/CTA: music reaches resolution. Satisfying harmonic landing.
Not a full swell — social proof videos are not trailers — but a complete,
settled ending. The viewer should feel something has been concluded.
Genre guidance:
- B2B/SaaS: minimal piano or guitar, light percussion, professional and warm
- Consumer product: acoustic guitar, light pop production, relatable and optimistic
- Health/wellness: organic instruments (piano, strings), calm and reassuring
- E-commerce/fashion: upbeat but sophisticated, 105-120 BPM, modern production"
Complete Example Prompts
Example 1: B2B Case Study — The Result-First Format (15s)
SEEDANCE 2.0 PROMPT:
Black frame. Single number fades in center: "312%". Large, white, bold
sans-serif. Below it in smaller text, also fading in 0.5s later: "increase
in qualified leads. 90 days."
0-2s: Number holds. Camera slowly pushes in toward the number — subtle,
barely perceptible movement. Sound: deep, resonant single note. Ambient
silence otherwise. The number is given space to land.
2-3s: Dissolve transition. Number dissolves into a face — the customer.
@material[image1] used as reference for customer portrait or as direct image source.
Subject seated slightly right of center in warm interview framing, looking
left-of-camera. Background: soft-bokeh office environment. Lighting: warm
three-point interview setup, key light from left, slight rim light separating
from background.
3-7s: Customer begins speaking. Camera is static. Every word is clear.
Warm room tone underneath voice. Below subject, lower-third text appears
at 4s: "[Customer Name] — [Title], [Company]". Clean sans-serif, white,
small. Above the lower-third, no other graphics yet — let the face tell
the story.
Sound: minimal music bed enters at 3s — -20dB under voice. Piano and
light percussion. Warm, confident. Does not compete with voice.
7-10s: Key quote moment. The most powerful phrase the customer says. At
this exact point, text appears over or beside the customer — the quote
itself in large typography. White, bold, appears word by word with 0.1s
stagger. Camera micro push-in: 1% zoom over 3 seconds — imperceptible
but felt. Music bed adds second element (pad or second instrument) at 7s.
10-13s: Camera pulls back slightly. Product screenshot or UI appears in
background on desk or screen, associating the result with the product.
@material[image2] used as product UI reference. Soft focus on product UI — it is
present but the customer remains the focal point.
Sound: music becomes slightly fuller. Voice continues or concludes.
Ambient environment enriches — background becomes more alive.
13-15s: Final frame. Customer finishes speaking. Five-star rating appears
below lower-third — animated left to right, 0.15s per star stagger, gold.
Platform attribution beneath stars. Product logo in upper-right corner,
small, fades in at 14s. Music resolves on final second. Warm, complete.
Material references: @material[image1] for customer photo, @material[image2] for product
screenshot, @material[audio1] for customer voice recording if available.
Example 2: Consumer Review — The Emotion Capture (10s)
SEEDANCE 2.0 PROMPT:
Open on close-up of a face mid-expression — a genuine, warm smile.
Eyes slightly crinkled. The expression is specific, not posed. Tight
framing: face fills 70% of frame. Background: soft bokeh, warm residential
environment (living room, kitchen — identifiable but indistinct).
0-1.5s: Face holds in close-up. No words yet. Sound: warm room tone,
faint background ambience of a comfortable home. Light music begins —
acoustic guitar, quiet, barely audible. The face communicates everything
before language.
1.5s: Camera pulls back slowly. Person becomes visible in context —
seated comfortably, the product visible in their hands or on the surface
near them. @material[image1] used as product reference. Natural, candid framing.
Documentary natural lighting treatment: window light from one side,
slight asymmetry, real and earned.
2-5s: Person speaks. Voice is clear, slightly warm in audio treatment.
Room tone underneath. They are sharing a specific moment — not a general
endorsement. The specificity is the credibility.
At 3s: first text overlay appears — a short fragment of what they're saying,
pulled into large typography. 3-4 words max. Bold, white, positioned lower-
third but large — almost subtitle-sized but treated as a design element.
5-7s: Product comes into focus. Camera drifts slightly to include product
more centrally. Person's hands interact with it naturally. Light catches
the product. @material[image1] product reference at this framing moment. The product
is shown in use, in life — not in a studio.
Sound: music bed grows subtly +3dB from 5-7s. Not louder, richer — a second
guitar line enters. Signals the story is reaching its point.
7-9s: Quote completes. Final sentence lands. At the moment the key word
lands — the result, the feeling, the reason — a second graphic element
appears: star rating (5 stars, gold, staggered entrance) and product name
below. Camera finishes its drift, composition settles.
9-10s: Hold on the settled frame. Music resolves. Room tone continues.
Warm, complete, real.
Material references: @material[image1] for product (for visual reference/integration).
@material[audio1] for customer audio if available. @material[image2] for product logo.
Example 3: SaaS Case Study — The Data Visualization (12s)
SEEDANCE 2.0 PROMPT:
Dark background, near-black with slight warm undertone — #0D0B09.
Premium data environment. Single line chart, minimal axes, centered frame.
0-1s: Empty chart axes appear — clean white lines, thin, on dark background.
A single label: "Monthly Recurring Revenue". Sound: soft ambient electronic,
deep and slow. A single ambient tone — held, expectant.
1-4s: Chart line begins building from left. It moves slowly, nearly flat —
the plateau before the change. Color: muted blue-gray, #8B9DC3. This is
the "before" period. The flatness is visible and intentional. Camera: static.
Sound: no change. Flat and honest about what was happening.
At exactly 4s: vertical dotted line appears on the chart — "Started using
[Product]". Text label appears at the top of this marker line in small caps.
Sound: a subtle, quiet tone — not dramatic, just a marker. A chime, brief.
4-7s: From the marker line, the chart diverges upward. New color: warm gold
#F5A623. The line climbs. Not gradually — it accelerates. Camera begins
a slow push-in toward the chart as it climbs. Sound: ascending ambient tone,
very subtle. The musical harmonic rises with the line. Not heavy-handed —
felt more than heard.
At the endpoint of the line (7s): the final value appears — a bold number
floating adjacent to the chart endpoint. Large, white, clear. Below it in
smaller text: the timeframe. @material[image1] if a real chart/data visualization
is provided as reference.
7-9s: Hold on the chart with the result visible. Camera rests. Sound:
music bed settles at this moment — a two-bar loop, warm, clean. Below
the chart, customer attribution fades in: "[Company Name]", small logo
mark if available via @material[image2]. Customer photo in small circle avatar
to left of attribution.
9-11s: Customer voice enters (if available via @material[audio1]). One or two
sentences — the human confirmation of the chart's story. A single
quote line appears in large typography, one phrase, above the chart.
Bold, white. The data told the story; the voice confirms it.
11-12s: Product logo appears center-right. Chart and quote remain visible.
Music resolves. Final composition: data, attribution, product logo — three
elements of proof in one clean frame.
Material references: @material[image1] for actual data chart or graph as visual
reference. @material[image2] for company/customer logo. @material[audio1] for customer voice.
Example 4: Review Montage — The Flood of Proof (8s)
SEEDANCE 2.0 PROMPT:
Clean white or very light gray background. Premium, minimal.
0-1s: Single large number appears center-frame, counting up rapidly.
The aggregate: "4,847 reviews". Counter accelerates and decelerates into
final number. Bold, dark typography on light background. Below it,
five gold stars in a horizontal row — fade-in together, no stagger here,
they arrive as a unit. Sound: rapid counting ticks, resolving into
a satisfying two-note landing as counter stops.
1s: Hard cut. First customer face — tight portrait. Warm, documentary
lighting. Off-camera look. 0.8s hold. Below face: name, city or role.
Camera: slight push-in. Sound: ambient warmth. Brief audio slice of
their voice — one word or half a second of speech.
1.8s: Hard cut. First result metric. Bold number, large, center-frame.
Dark background (complement to light-background face shots — alternating
treatment). "47 hours saved per month." Number animates in with impact.
Camera: static. Sound: single bass note thud on impact.
2.6s: Cut. Second face. Different demographic — different gender, age,
context than first. Same lighting treatment, same shot structure —
consistency signals intentionality. Off-camera look, warm light. 0.8s.
Audio slice: one-word or brief exclamation.
3.4s: Cut. Second result metric. Different color treatment, same bold
typography. "3x conversion rate." Same impact animation. Sound: same
bass note — creates rhythmic expectation, viewer anticipates each cut.
4.2s: Cut. Third face. Wide shot this time — a bit more environment visible.
Person at their workspace. The context is meaningful.
4.8s: Cut. Third metric — this time a star rating visual with the score.
"4.9 / 5.0" with star cluster. More complex graphic, slightly longer hold.
5.8s: Cut. Montage accelerates — two more rapid face/metric pairs.
0.4s each. Music track (entering at 4s at -18dB) becomes audible here,
riding the rhythm of the cuts. Faces and numbers intercut like heartbeats.
7s: Hard cut to black. Single frame. Sound: everything stops.
7-8s: Slow fade in: product logo, centered, white on dark. Tagline appears
below. "Join 4,847 customers." Music final note sustains and releases.
Clean, complete.
Material references: @material[image1]-@material[image6] for customer photos (different people).
@material[image7] for product logo. All faces should represent diversity in age,
gender, and context.
Show full SKILL.md (343 more words)Show less
Prompt Rules for Testimonial Content
Lead with the result, not the person. Viewers decide in the first two seconds whether to trust what follows. Give them the evidence first — the number, the metric, the outcome — then introduce the human source.
Make every quote earn its screen time. A quote that lasts longer than 3 seconds needs to work visually as hard as it works verbally. Pair it with motion, data, or complementary imagery. Never let static text sit on a static background for more than 2 seconds.
Specificity is the currency of credibility. "I doubled my revenue" is weaker than "We went from $12K to $31K MRR in 11 weeks." Prompt for specific numbers, timeframes, and named outcomes wherever possible.
Customer voice is the primary audio signal. If you have a voice recording, it goes first in the mix. Music and sound effects serve the voice; they do not share equal footing with it. A viewer who cannot clearly hear the customer will not trust the testimonial.
Match the visual register to the audience. B2B testimonials require authority signals: clean environments, professional lighting, data visualization. Consumer testimonials require relatability signals: real spaces, natural light, organic moments. Never use a studio treatment for a consumer brand, or a casual treatment for an enterprise pitch.
Star ratings need a source. Five stars with no platform attribution reads as fabricated. Always include the platform name — even small, even subtle. The attribution is the proof behind the proof.
The final frame is a commitment device. The last image the viewer sees before deciding to act should be: product in context, or the customer's face with the result attribution, or the aggregate social proof signal. Never end on a generic call-to-action graphic.
Avoid the endorsement pose. A customer facing directly into the camera with a wide smile, speaking scripted praise, is the exact visual language of advertising everyone has learned to distrust. Aim for slightly off-camera, genuine expression, specific language. If it looks like a testimonial, it performs like one.
Seedance Testimonial Story next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
Seedance Testimonial Story compared with similar skills
Skill
Stars
Used in
Tokens
Auto-check
Licence
Repo updated
Seedance Testimonial Story this skillrediumvex/ai-video-generator-claude
This skill should be used when the user has a vague Seedance 2.0 video idea and asks for creative guidance, story development, scene planning, a director interview, or help turning an undeveloped…
This skill should be used when the user asks for Spanish Seedance 2.0 prompt wording, Spanish cinematic vocabulary, or translation of camera, lighting, action, VFX, audio, and production terms into…
This skill should be used when the user asks for Japanese Seedance 2.0 prompt wording, Japanese cinematic vocabulary, or translation of camera, lighting, action, VFX, audio, and production terms…
This skill should be used when the user asks for Korean Seedance 2.0 prompt wording, Korean cinematic vocabulary, or translation of camera, lighting, action, VFX, audio, and production terms into…
This skill should be used when the user asks for Russian Seedance 2.0 prompt wording, Russian cinematic vocabulary, or translation of camera, lighting, action, VFX, audio, and production terms into…
Generate customer testimonial and social proof video prompts for Seedance 2.0 on Higgsfield. Seedance Testimonial Story is an agent skill from rediumvex/ai-video-generator-claude.0 on Higgsfield.
When should I use Seedance Testimonial Story?
Seedance Testimonial Story fits situations like: customer stories; case study videos; review showcases; social proof content.
How do I install Seedance Testimonial Story in Claude Code?
Run `npx skills add rediumvex/ai-video-generator-claude --skill seedance-testimonial-story -a claude-code`. Or copy the skill folder (skills/08-testimonial-story in rediumvex/ai-video-generator-claude) into .claude/skills/seedance-testimonial-story in your project. Claude Code loads it when a task matches its description.
How do I install Seedance Testimonial Story in Codex?
Run `npx skills add rediumvex/ai-video-generator-claude --skill seedance-testimonial-story -a codex`. Or copy the skill folder (skills/08-testimonial-story in rediumvex/ai-video-generator-claude) into .agents/skills/seedance-testimonial-story in your project. Codex loads it when a task matches its description.
Can I use Seedance Testimonial Story in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add rediumvex/ai-video-generator-claude --skill seedance-testimonial-story -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/seedance-testimonial-story, .gemini/skills/seedance-testimonial-story, .github/skills/seedance-testimonial-story and .opencode/skills/seedance-testimonial-story in your project.
What does Seedance Testimonial Story need to run?
Going by SKILL.md and its folder, Seedance Testimonial Story needs the command-line tools its instructions call (black).
Does Seedance Testimonial Story access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Seedance Testimonial Story safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Seedance Testimonial Story use?
Seedance Testimonial Story is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Seedance Testimonial Story use?
About 7.2k tokens (SKILL.md is roughly 29k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Seedance Testimonial Story?
Skills that share tags, products or a category with Seedance Testimonial Story: Seedance Interview (1370998960-del/https-github.com-Emily2040-seedance-2.0, 241 stars), Seedance Brand Story (beshuaxian/higgsfield-seedance2-jineng, 952 stars), Seedance Vocab Es (Emily2040/seedance-2.0, 7.6k stars) and Seedance Vocab Ja (Emily2040/seedance-2.0, 7.6k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Seedance Testimonial Story?
rediumvex (a GitHub user) maintains it in rediumvex/ai-video-generator-claude, which has 409 GitHub stars. The repository holds 10 skills in this directory. The repository was last updated on August 11, 2026.