create-provider-package-with-core

Build custom provider adapters on @prefactor/core for agent, LLM, and tool tracing.

3|Updated Jan 20, 2026
One-click install
npx skills add https://github.com/prefactordev/typescript-sdk --skill create-provider-package-with-core
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: create-provider-package-with-core
Source: https://github.com/prefactordev/typescript-sdk/tree/main/skills/create-provider-package-with-core
Command: npx skills add https://github.com/prefactordev/typescript-sdk --skill create-provider-package-with-core

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Use when building a custom provider integration on top of @prefactor/core so your app can instrument agent, llm, and tool workflows without relying on a prebuilt adapter package.

Core Features & Use Cases

  • Define adapter boundaries and reuse core primitives for tracing.
  • Build thin provider wrappers that map provider APIs to core instrumentation.
  • Ship internal tooling or private SDKs with consistent telemetry across providers.

Quick Start

Explain how to bootstrap a custom provider adapter that uses @prefactor/core for instrumentation in your app.

Frequently Asked Questions about create-provider-package-with-core

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

FAQPage Schema
How do I build a custom provider adapter for TypeScript LLM and tool workflows?

To build a custom provider adapter for TypeScript workflows, you wrap provider APIs and map them to core instrumentation primitives, ensuring consistent tracing across agents, LLMs, and tools without relying on prebuilt packages.

When do I need to instrument a custom provider integration on top of core primitives?

You need to instrument a custom provider integration on core primitives when your application requires consistent telemetry and tracing for agent, LLM, or tool workflows that prebuilt adapter packages do not support.

What's the best way to define adapter boundaries for LLM provider tracing?

The best way to define adapter boundaries for LLM provider tracing is to build thin provider wrappers that map external provider APIs directly to core instrumentation primitives, isolating provider-specific logic from core telemetry.

Does this approach handle error handling and provider-specific spans for agent instrumentation?

Yes, this approach handles error handling and provider-specific spans for agent instrumentation by leveraging core primitives to maintain consistent tracing and production readiness across custom provider integrations.

Can I use this method to ship an internal SDK with consistent telemetry across providers?

You can use this method to ship internal tooling or private SDKs with consistent telemetry across providers by mapping diverse provider APIs to unified core instrumentation primitives for reliable tracing.

What are the limitations of building custom provider adapters instead of using prebuilt packages?

Building custom provider adapters instead of using prebuilt packages requires you to manually manage adapter boundaries, provider-specific spans, and error handling, increasing maintenance overhead for production readiness.