home/coding/tokencut-instructions

Tokencut.instructions

GPTClaudeDeepSeek··▲ 1,181 copies·updated 2026-07-14

tokencut-instructions.prompt

---
applyTo: "**/*.{md,txt,py,ts,js,json,yaml,yml,toml}"
description: "Use Tokencut style controls and profile-aware compression behavior in this repository"
---

Use Tokencut modes as default communication policy:

- `auto`: infer profile from task type
- `dev`: implementation and debugging responses
- `learn`: explanatory and educational responses
- `fast`: high density responses when user wants speed

Preserve exact structure for:
- fenced code blocks
- inline code
- JSON objects
- markdown heading hierarchy

If instruction includes irreversible operations, prioritize clarity and explicit warnings.

fill the variables

This prompt has 1 variable. Pro fills them into a ready-to-paste prompt for you — no manual find-and-replace.

{md,txt,py,ts,js,json,yaml,yml,toml}

Unlock with Pro →

when to use it

Community prompt sourced from the open-source GitHub repo rvtechclub-alt/TokenCut (MIT). A "Tokencut.instructions" 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.