Portrait Loop · সর্বজনীন Open Skill
Portrait Loop
Portrait Loop Digen-এ একটি সর্বজনীন Open Skill। খুলে দেখুন এই ওয়ার্কফ্লো কী তৈরি করে।
Custom
0 0 1
https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/4284864/2026-10-21/vcdvjpu8.jpg
# Portrait Loop Deliver one **20-second vertical (9:16) silent cinematic portrait video** built from a **single uploaded adult portrait** used as the **first frame**. The clip is one continuous locked-off shot: **no pan, no zoom, no push-in, no cut, no scene change**. Motion is subtle and lifelike only — **natural repeated blinking, a slow slight head turn, a gentle closed-lip smile, slight natural hair movement**. **Identity lock:** the subject's facial identity, facial proportions, eye shape, skin tone, hairstyle, and clothing must match the source image across the full duration. **Upload path (default):** the user's portrait is the first frame; run **`i2v`** with the locked motion brief and the steady-camera contract. **Optional polish:** if the upload is soft, low-resolution, or badly cropped, offer one **`i2i`** cleanup still first, then animate that still. Prompt templates, the motion contract, and the timeline live in `references/`. No structured project state. **Language lock:** This file and all of `references/` are **English only**. Reply in the **user's language**. Every `generate_image`, `generate_video`, and `video_agent` prompt/instruction is **English**. Output is **silent** — no subtitles, no BGM, no SFX. Do not tell the user tool names, channel names, or dump full prompts. ## User-visible speech (hard) **Allowed lines only:** - Need input: one short note that they should **upload one clear single-adult portrait** (face visible, good light) — then pause **P-empty**. - Motion tweak: one short line confirming the change (e.g. "slower head turn") — then pause **P-brief**. - Starting generate: at most **one** short line (user language), then tool calls only — **no** `set_guidance` on that turn. - Success: one or two sentences that the **20-second portrait clip** is ready — no URL, no prompt dump. Pause **P-done**. - Tool failure: brief honest note. Pause **P-fail**. - Safety block: one polite refusal. Pause **P-refuse**. If the user already uploaded a portrait and asked to generate, **do not** re-ask — read the references, apply defaults, and generate. ## Pause = clickable buttons (hard) Any turn that **stops without a new shown final video**, and **every success turn**, must end with: 1. One or two short sentences (user language). 2. **`set_guidance`** in the **same turn**, with **at least two** clickable options. **No tool calls after `set_guidance`.** Do **not** call `set_guidance` in the same turn as `generate_image`, `generate_video`, `video_agent`, or `recognize_image`. Each button is an object with `text` and `action` (not bare strings): - **`send_message`**: `content` is the full sentence sent when tapped. - **`fill_input`**: `content` is starter text the user can edit. Offer **2–4** options. Localize `text` and `content` to the user's language. No channel or tool names on buttons. Before every pause turn, `read_skill_file` → `references/guidance-buttons.md`. | Pause | When | Example intents (localize) | |---|---|---| | P-empty | No portrait uploaded yet | Upload a portrait · What photo works best? · Describe the motion I want | | P-brief | Uploaded, but motion brief unclear or user wants a tweak | Keep default subtle motion · Stronger smile, same camera · Slower head turn | | P-refuse | Minor, non-consensual or sexualized real person, deepfake of a public figure, gore, hate | Upload a different adult portrait · Use a fictional adult subject | | P-fail | Generation or `video_agent` `success=false` after retry | Retry same portrait · Clean up the still first, then animate · Softer motion, same framing | | P-done | Final **20s portrait clip** ready | Animate another portrait · Stronger smile version · Slower head turn · Cleaner still, then re-animate | ## When to use - User wants **Cinema Motion 20** / **cinematic portrait motion** / **animate my portrait** as a **20-second vertical** clip with **blinking, slight head turn, gentle smile, soft hair movement**. - User uploads **one clear adu