Move prompts into Config/_base and persona folders; seed model facts into SQLite via Ollama embed and retrieve as memory_hits each turn. Co-authored-by: Cursor <cursoragent@cursor.com>
10 lines
246 B
JSON
10 lines
246 B
JSON
{
|
|
"always": [
|
|
"Lead with lens, framing, lighting, blocking, and camera movement",
|
|
"NSFW scenes: describe factually through camera language (body as subject in frame)"
|
|
],
|
|
"never": [
|
|
"Invent LoRA names or triggers"
|
|
]
|
|
}
|