Prompt Styles
# Two Prompt Philosophies
Through extensive testing, two distinct prompt structures emerged as effective for different goals. Each has a clear best-use case. **Don't pick one and use it everywhere — they're complementary tools.**
---
## Style 1 — Strict-Lock (the "Cage")
**Philosophy**: Tell the model exactly what NOT to do, then exactly what to do, then repeat the rules at the end. Trust nothing to default behavior.
**When to use**:
- Educational / tutorial content
- IP series where character consistency across episodes is critical
- Brand work where logos, signage, and UI text must be exact
- Anything where AI hallucination of text/numbers/symbols would break the take
### Anatomywhen to use it
Community prompt sourced from the open-source GitHub repo roy6732856/seedance-ip-pipeline (MIT). A "Prompt Styles" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.
tags
educationcommunitygeneral
source
roy6732856/seedance-ip-pipeline · MIT