Your editing preferences — short sentences, six-word headlines, em-dashes intact — declared across a chat in two languages, preserved as standing rules.
Personal voice preferences are the most repeated and least preserved instructions in AI work: every new session re-teaches the same rhythm, the same headline length, the same punctuation rules. This distillation captures the declarations — including the "from now on" and "final version" markers that signal standing rules — and the detector reads correction language in English AND Turkish: "böyle değil", "şöyle yap", "bunu koru", "bundan sonra", "son hali" all register as the revision signals they are. This setup loads a bilingual editing session whose preferences become a permanent rule set.
Declare in your language
The detector reads revision signals in English and Turkish — "bunu koru" counts exactly like "keep this".
Mark the standing rules
"Bundan sonra" and "from now on" flag rules that outlive the session — they get the [standing rule] tag.
Paste once per session
The preference prompt opens each session; the re-teaching tax is gone.
It reads revision language in both. The detector treats "bunu koru" like "keep this", and "bundan sonra"/"son hali" flag standing rules the same as "from now on". In the loaded example those become tagged [standing rule] requirements, while a rejected line, "böyle değil", lands on the AVOID list. The conversation-to-prompt-builder extracts these VERBATIM; read the prompt once before trusting a real run.
The REQUIREMENTS block is labelled "later items override earlier ones where they conflict", so the last decision wins, the way the conversation actually resolved: a six-word headline rule stated later beats an earlier looser one. Directions you walked back travel on the AVOID list so the model can't rediscover them. The conversation-to-prompt-builder distills this into a reusable prompt you paste at the start of each new session.
What a converged conversation contains: a rejected voice, an accepted framing, a hard constraint, a format decision, and a final-version moment — each mapped to its prompt section.
The result you finally liked lives inside the chat. Distill the iterations into one prompt: corrections become requirements, the accepted version becomes the quality bar.
In feedback threads the rejections carry the knowledge: the mocking empty-state enthusiasm, the passive phrasing — converted into the avoid-rules that define the style.
End a working session like a shift change, not an abandonment: state captured, decisions logged, next step named — ready for the next session to pick up.
Turns a reusable API request prompt into a clean variable set covering endpoint, method, auth, payload, response format, and error handling.
A reusable AI agent task template with variables for objective, context, available tools, constraints, success criteria, failure handling, and output format.
Distill an iterated conversation into one reusable prompt — corrections become requirements, rejections become an avoid-list.
AI nails your brand voice on one output, then slides back to generic hype on the next — banned words, wrong tone, dropped format. Here's how to turn your voice into rules, a banned list, examples, and a checklist the model can actually follow.