reasoning-models
Multi-provider reasoning/thinking configuration. Discriminated unions for OpenAI effort, Anthropic budget, Google level/budget, DeepSeek middleware. Triggers on "reasoning", "thinking", "o1", "deepseek", "extended-thinking", "thinkingEffort", "reasoningEffort".
npx skills add majiayu000/claude-skill-registry --skill reasoning-models --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# Reasoning Models Multi-provider reasoning system with discriminated union configs. Each provider has different API: OpenAI effort levels, Anthropic token budgets, Google level/budget, DeepSeek tag extraction. Registry pattern for handlers. ## Discriminated Union Types One type per provider - TypeScript enforces which fields are valid: ```typescript // From apps/web/src/lib/ai/reasoning/types.ts export type ReasoningConfig = | { type: "openai-reasoning-effort"; effortMapping: Record<ActiveThinkingEffort, string>; summaryLevel?: "brief" | "detailed"; useResponsesAPI: boolean; } | { type: "anthropic-extended-thinking"; budgetMapping: Record<ActiveThinkingEffort, number>; // Token budgets betaHeader: string; // e.g., "interleaved-thinking-2025-05-14" } | { type: "google-thinking-level"; // Gemini 3 family levelMapping: Record<ActiveThinkingEffort, "low" | "medium" | "high">; includeThoughts: boolean; } | { type: "google-thinking-budget"; // Gemini 2.5 budgetMapping: Record<ActiveThinkingEffort, number>; } | { type: "deepseek-tag-extraction"; tagName: string; // e.g., "think" applyMiddleware: true; } | { type: "generic-reasoning-effort"; // xAI, Perplexity, Groq parameterName: string;
- Discriminated Union Types
- Effort Levels
- Model Config Integration
- buildReasoningOptions
- Handler Registry Pattern
- Provider-Specific Handlers
- OpenAI - Reasoning Effort
- Anthropic - Extended Thinking Budget
- Google - Thinking Level/Budget
- DeepSeek - Tag Extraction Middleware
- Generic - Simple Effort Parameter
- Provider Options Output
- Key Files
- Avoid
What does the reasoning-models skill do?
Multi-provider reasoning/thinking configuration. Discriminated unions for OpenAI effort, Anthropic budget, Google level/budget, DeepSeek middleware. Triggers on "reasoning", "thinking", "o1", "deepseek", "extended-thinking", "thinkingEffort", "reasoningEffort".
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill reasoning-models --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.