Prompt Engineering

Preserve Prompt Preferences — Voice Rules That Stick

Your editing preferences — short sentences, six-word headlines, em-dashes intact — declared across a chat in two languages, preserved as standing rules.

Overview

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.

Workflow

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.

Why This Works

Standing-rule markers separate session feedback from permanent preference
Bilingual detection meets the conversation where it happened
Verbatim preservation keeps the preferences in your own words

Best for

Writers with strong, specific voices
Editing collaborations that converge on rhythm rules
Anyone re-teaching preferences every session

Not for

Mid-project session continuity — the Context Handoff Builder carries working state
Preferences the AI should know across ALL projects permanently — the future Project Context Builder

Use cases

Bottling voice rules taught across sessions
Preserving "from now on" declarations
Bilingual iteration sessions (EN/TR signals)

Customize This Resource

Opens this scenario in Conversation-to-Prompt Builder. Distill to get the reusable prompt — then adjust the output mode.

Open in Conversation-to-Prompt Builder

Prompt Template

REUSABLE PROMPT
--------------------------------
TASK
Edit my draft announcements in my voice.

REQUIREMENTS (earned through iteration — later items override earlier ones where they conflict)
- Şöyle yap: keep my short sentences, even the fragments.
- Bu daha iyi. Bunu koru — the rhythm is the voice.
- Bundan sonra başlıkları en fazla altı kelime tut. [standing rule]
- Harika olmuş. Son hali bu — tam istediğim. [standing rule]

CONSTRAINTS
- [No hard constraints detected — add any that must hold]

FORMAT
- Edit this product announcement draft in my voice.

AVOID (rejected during the conversation — do not reproduce these)
- Böyle değil — that formal transition style is exactly what I avoid.

QUALITY BAR
Match the version the conversation accepted — the reaction was: "Bu daha iyi. Bunu koru — the rhythm is the voice."
If anything above is ambiguous or conflicting, say so and ask — do not silently pick one reading.
--------------------------------

WHY THIS PROMPT WORKS
- It starts where the conversation ENDED: the corrections and accepted preferences are requirements now, not future feedback.
- The last decision wins: the prompt states that later requirements override earlier ones, the way the conversation actually resolved.
- Rejected directions travel as an AVOID list — the model cannot rediscover the paths you already walked back.

PRESERVED REQUIREMENTS
- Corrections carried: 1 · Accepted preferences: 1 · Standing rules: 2
- Constraints: 0 · Format decisions: 1

REJECTED DIRECTIONS
- Böyle değil — that formal transition style is exactly what I avoid.

OPTIONAL VARIABLES
- None placed. Switch to Template Prompt mode for an {{input}} slot, or variableize fully in the Prompt Variable Builder.

USAGE NOTES
- Paste this prompt at the start of a new conversation to get the converged result without re-running the iterations.
- Adjust the task line for each use; the requirements below it are the part the iterations earned.
- Distillation confidence: High.
- Signals were extracted VERBATIM from the conversation (deduplicated, never paraphrased) — read the prompt once before trusting it with a real run.

Related Resources

Explore more resources that work well with this one.

More from Conversation-to-Prompt Builder

Convert a Conversation to a Prompt — the Full Anatomy Turn a Chat Into a Prompt — Start at the Result Convert a Feedback Thread to a Prompt — Rejections Are the Asset

View all Conversation-to-Prompt Builder resources →

Adjacent Workflows

AI Session Handoff — Shift Change for Working Sessions Context Handoff Builder API Request Variable Builder Prompt Variable Builder Agent Task Template Prompt Template Builder

Related Tools

Conversation-to-Prompt Builder

Tip: Save time by exploring related resources and tools that integrate with this workflow.

Explore all resources