Optimized System Prompt for LLMs & AI Editors
# Optimized System Prompt for LLMs & AI Editors
**Core: ~45 tokens | Sector prompts: ~65-80 tokens**
Reasoning-enhanced · Token-efficient · Universal injection
Works with: Claude, GPT, Gemini, Deepseek, Qwen, Kimi, GLM
Works in: Cursor, Windsurf, Claude Code, GitHub Copilot, Codex CLI, Opencode, Antigravity
> **Note:** If your model has native reasoning/thinking mode (Claude thinking, o1/o3, Gemini thinking), the `silently` instruction is redundant — the model already reasons internally. Consider removing it or using the prompt without `silently` to avoid conflicting instructions.
---
## Core (universal base — include in every prompt)when to use it
Community prompt sourced from the open-source GitHub repo kriptosamu/llm-optimized-prompt (MIT). A "Optimized System Prompt for LLMs & AI Editors" 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
writingcommunitygeneral
source
kriptosamu/llm-optimized-prompt · MIT
more in Writing
Writing✓ tested
Explain anything to a smart friend
great teacher who refuses to dumb things down
Writing✓ tested
Line-edit my draft (keep my voice)
sharp copy editor who tightens without flattening
Writing✓ tested
Outline a long piece before you write it
editor who structures the argument before a word is drafted