mehbylabs/creator-skills / skills/creator-character
creator-character
FlaggedKeep a person, mascot, or character consistent across many images and videos with whatever providers the user has. Builds a reusable character pack (reference sheet, anchor description, approved reference images) and reuses it in every generation; uses a Higgsfield Soul ID when available. Use when the user wants the same face or character in multiple images, a recurring mascot, a consistent AI influencer or presenter, "use my face", a character sheet, or a story with recurring characters. Not for one-off images (creator-image).
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 3.3 KB |
| references/arabic-voice.md | markdown | 7.7 KB |
| references/audio.md | markdown | 3.4 KB |
| references/browser-aistudio-browser.md | markdown | 2.8 KB |
| references/browser-flow-browser.md | markdown | 4.5 KB |
| references/browser-gemini-browser.md | markdown | 4.8 KB |
| references/config.md | markdown | 2.5 KB |
| references/connect.md | markdown | 3.9 KB |
| references/image-prompting.md | markdown | 1.7 KB |
| references/media-ops.md | markdown | 2.4 KB |
| references/providers.md | markdown | 7.0 KB |
| references/video-prompting.md | markdown | 1.9 KB |
| scripts/connect.pyexecutable | script | 16.6 KB |
| scripts/creator.pyexecutable | script | 71.1 KB |
| scripts/media.pyexecutable | script | 15.7 KB |
references/audio.md
View on GitHub# Audio: voiceover, music, sound effects
## Voiceover (`creator.py speech`)
```bash
python3 scripts/creator.py speech --text-file script.txt --out vo.mp3 [--lang ar-EG] [--voice ID] [--style "warm, upbeat"] [--speed 1.05]
```
**Arabic or any Arabic dialect → read `arabic-voice.md` first** (script must be written in
the dialect; `--lang` picks dialect-aware routing and voices). **Add tashkeel to every word
whose meaning depends on its vowels before sending it to any TTS**; `speech` refuses Arabic
text with such words left bare (exit `5`), and `creator.py tashkeel-check --text-file F`
lists them ahead of time. Add `--verify` to transcribe the result back and list words the
voice read differently from the script (`verify` block in the JSON; `creator.py verify` checks
an existing file). Use it on every line of a long or fact-heavy script.
- Write for the ear: short sentences, numbers as words when ambiguous ("twenty twenty-six"),
spell out acronyms you want letter-by-letter ("A-P-I").
- Long scripts: split into one file per scene/paragraph so retakes are cheap and the
explainer can time scenes to audio.
- `--style` is delivery direction. Google TTS and OpenAI `gpt-4o-mini-tts` honour it;
ElevenLabs ignores it (choose the voice instead; `eleven_v3` reads inline tags like
`[whispers]`, `[laughs]`). Google also reads inline tags: `<short pause>`, `<laugh>`, `<sigh>`.
Voices:
| Provider | Default | Good alternatives |
|---|---|---|
| ElevenLabs | first voice in the account (set `voices.elevenlabs` in config) | list with `GET /v2/voices`; the legacy default voices retire Dec 31 2026 |
| Google | `Kore` | `Puck` (upbeat), `Charon` (deep, informative), `Zephyr` (bright), `Aoede`, `Fenrir`, `Leda`, `Orus`, `Enceladus`, `Sulafat` |
| OpenAI | `marin` | `cedar`, `coral`, `sage`, `ash`, `nova`, `onyx` |
| Google Cloud TTS | `Kore` | same names as Gemini; locale from `--lang` |
| Azure | per `--lang` country voice (e.g. `ar-EG-SalmaNeural`) | any `<locale>-<Name>Neural` |
Use the same provider + voice for every line of one project. If fallback switched provider
mid-project, regenerate the earlier lines on the new provider rather than mixing voices.
## Music (`creator.py music`)
```bash
python3 scripts/creator.py music --prompt "lo-fi hip hop, warm Rhodes, vinyl crackle, 80 bpm" --duration 60 --instrumental --out bed.mp3
```
Prompt with genre, instrumentation, tempo (bpm), mood, energy arc ("builds at 0:20").
Background beds under narration: always `--instrumental`, mixed by `media.py assemble`
(ducked automatically). Google's default `lyria-3-clip-preview` returns 30 s; pick
`lyria-3.5` in config for full-length tracks.
The router checks every track on disk: silent or truncated takes are thrown away and
retried; a track shorter than `--duration` (or over a provider cap, e.g. ElevenLabs 300 s)
is extended seamlessly to the exact length (`notes` in the JSON). Still listen once before
shipping: a check can prove a file has sound, not that the sound fits.
## Sound effects (`creator.py sfx`)
```bash
python3 scripts/creator.py sfx --prompt "whoosh transition, airy, fast" --duration 1.2 --out whoosh.mp3
```
Describe source, action, space, and length: "heavy wooden door creaks open in a stone hall".
`--loop` for ambiences.
## Licensing
Generated audio is subject to the provider's terms and the user's plan; mention this when
the user says the piece is for paid ads or broadcast.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |