speculative-naming

Standardize speculative decoding identifier names in the SGLang codebase.

Updated Jun 4, 2026
One-click install
npx skills add https://github.com/annealing-inversion/sglang-kimi-deferral --skill speculative-naming
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: speculative-naming
Source: https://github.com/annealing-inversion/sglang-kimi-deferral/tree/main/.claude/skills/speculative-naming
Command: npx skills add https://github.com/annealing-inversion/sglang-kimi-deferral --skill speculative-naming

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Naming conventions for SGLang speculative decoding identifiers to reduce ambiguity, enforce consistency, and improve code readability across the speculative decoding components and related backends.

Core Features & Use Cases

  • Consistency: Enforces a uniform naming scheme for speculative identifiers used in python/sglang/srt/speculative and related IPC fields, schedulers, and observability metrics.
  • Maintainability: Simplifies code reviews and refactoring by providing clear naming rules.
  • Use Case: When adding a new speculative identifier, rename an existing one to follow the standard convention, or review identifiers for consistency across the speculative module.

Quick Start

Rename or introduce a new speculative identifier to align with the established naming rules.

Frequently Asked Questions about speculative-naming

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I standardize speculative decoding identifiers in the SGLang codebase?

Yes, you can rename existing speculative decoding identifiers across schedulers, IPC fields, and observability metrics by applying scoped renaming practices and validating them against existing identifiers to improve code readability.

What naming conventions apply to SGLang speculative decoding bonus tokens?

Speculative naming guidelines enforce a uniform naming scheme for identifiers used in python/sglang/srt/speculative and related IPC fields, schedulers, and observability metrics to reduce ambiguity and improve code readability.

How do I review SGLang speculative identifiers for consistency across schedulers and backends?

Review SGLang speculative identifiers by validating them against existing identifiers and applying explicit naming rules across related backends, schedulers, and IPC fields to ensure consistency and maintainability.

Can I use speculative naming rules for CLI flags and observability metrics in SGLang?

Yes, speculative naming rules apply to CLI flags and observability metrics in SGLang, enforcing a uniform naming scheme and explicit conventions to simplify code reviews and refactoring across the speculative module.