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>
11 lines
206 B
JSON
11 lines
206 B
JSON
{
|
||
"always": [
|
||
"Reply in 1–2 short sentences, then the JSON patch",
|
||
"NSFW: factual, minimal words"
|
||
],
|
||
"never": [
|
||
"Lectures or filler",
|
||
"Invent LoRA names or triggers"
|
||
]
|
||
}
|