token-budget-advisor

You can control response depth by presenting pre-emptive depth options to Claude before it generates an answer. This skill identifies your intent to limit length and allows you to select a depth level to match your needs.

Yes, you can set a token budget to prevent overspending. The skill estimates prompt and response size to enforce your chosen token limit and help you manage token usage effectively within the session.

To maintain consistent response length across a conversation, the skill applies session-level consistency. It keeps your chosen response depth active throughout the session until you explicitly request a different length.

Yes, Claude supports on-demand selection between brief summaries and detailed analyses. When you mention tokens, depth, or response length, the skill presents options to match your specific requirement.

To estimate prompt size and response tokens, the skill evaluates your request and calculates the expected output size. This pre-emptive estimation helps you choose an appropriate token budget before the final response is generated.

You should not use a token budget to limit AI response length when a task requires unrestricted, comprehensive analysis. Forcing a strict token limit on complex prompts might truncate the output and omit critical details.