Configure extended thinking budgets for Claude API calls.
npx skills add https://github.com/markus41/claude --skill extended-thinking
Please help me install this Agent Skill. Skill: extended-thinking Source: https://github.com/markus41/claude/tree/main/.claude/skills/extended-thinking Command: npx skills add https://github.com/markus41/claude --skill extended-thinking
This Skill enables advanced reasoning capabilities within Claude, allowing it to perform complex, multi-step problem-solving and deep analysis that requires significant deliberation.
Use the extended-thinking skill to analyze this complex architecture decision with a budget of 10000 tokens.
High-intent search queries and answers about installing and using this skill.
Extended thinking is enabled for Claude API calls by activating the skill, which facilitates complex reasoning and multi-step problem-solving through configurable token budgets for internal deliberation.
The best way to analyze architecture decisions is using extended thinking to have Claude thoroughly deliberate on the options, providing deep analysis and recommendations based on a configurable thinking budget.
Yes, you can set specific token budgets for the thinking process to manage the depth of reasoning and control costs when using the Claude API for complex tasks.
Extended thinking supports various Claude models and provides guidance on tool use and best practices for carrying out complex tasks that require multi-step reasoning and deliberation.
You need extended deliberation when facing complex tasks requiring deep analysis and significant thought, such as software architecture decisions involving multiple trade-offs and multi-step problem-solving.
kylehughesCurate context and optimize prompts for Anthropic Claude models using XML structuring.
gkswls5006-webAutomate structured step-by-step reasoning for architecture decisions and complex analyses.
evanfang0054Enable chat models to invoke external tools and APIs with parallel execution.
pingdiorGuide multi-step internal reasoning and hypothesis verification before final responses.
