You are given a video of a person speaking to camera, plus the transcript of what
he says. Your job is a MOTION-GRAPHICS EDIT ONLY: add professional animated
graphics on top of the existing footage. Do not re-shoot, re-generate, or
re-interpret the person or the scene.

=== AUDIO LOCK (highest priority) ===
The attached video's voice track is the locked master. Keep it 100% untouched:
- Do not generate, replace, dub, revoice, clone, or re-time the speaker's voice
- Do not change pitch, speed, tone, or duration
- Do not add narration, voice-over, or TTS
- Do not add music of any kind
The only audio you may ADD is sound design: whooshes, soft clicks, pops, ticks,
subtle risers and impacts, timed to the graphics. Keep all SFX low in the mix so
the voice always stays clearly on top.

=== SUBJECT LOCK ===
Preserve the attached person exactly as filmed — identical face, facial features,
hair, skin tone, body, and clothing. Preserve the original background, framing,
and lighting. Keep his natural movement, his direct eye contact with the lens, and
his light expressive hand gestures as they are in the footage. His mouth movement
stays exactly as recorded and in sync with the original audio.

=== TRANSCRIPT (timing map — NOT to be spoken) ===
Below is what the audio already says. Use it only to know what is said at each
moment, so you can time every graphic to the exact word or phrase it represents.
Never speak, narrate, or re-record it.

" اكتب الاسكربت هنا "

=== MOTION GRAPHICS ===
Style: modern, punchy, premium YouTube-intro energy. Fast-paced but never cluttered.
While he speaks, rich and varied graphics animate in the space AROUND him:
- Attractively animated text and large keyword titles
- Icons, arrows, circles, squares, lines, and geometric shapes
- UI panels, cards, and callout boxes
- Modern snappy transitions between beats

Rules:
- Each element appears on the exact word or phrase it represents, then exits cleanly
- Never cover his face; keep graphics in the empty areas of the frame
- 1–2 focal elements at a time — build density through rhythm, not overlap
- Match the energy of his delivery: emphasis words get bigger, harder hits

=== ARABIC ON-SCREEN TEXT — STRICT ===
Do not generate any Arabic text from your imagination. Render ONLY the Arabic
strings I write between quotation marks, verbatim exactly as written:
right-to-left, with correctly connected cursive letterforms and correct contextual
letter shaping. No mirrored, reversed, broken, disconnected, or garbled glyphs.
No pseudo-Arabic. Use a clean bold Arabic display font, high contrast, fully legible.
If any string cannot be rendered perfectly, do NOT approximate it — replace it
immediately with a suitable non-text motion graphic instead: an icon, shape, arrow,
line, bar, or UI card.

=== LOGOS ===
Where the speech clearly refers to a specific real brand or product, you may show
its real logo, accurately and undistorted. If it cannot be rendered accurately,
use a generic icon instead.

=== TECHNICAL ===
16:9, high resolution, cinematic look, sharp focus on the subject, tidy soft
background. Output length must exactly match the attached video's length.

=== NEGATIVE ===
No music. No revoicing, TTS, dubbing, or audio replacement. No auto-generated
subtitles or burned-in captions unless explicitly listed above. No changes to face
or clothing. No watermarks. No Latin placeholder text. No gibberish or broken
Arabic. No distorted hands. No camera moves that break the original framing.