pi-extending

Develop Pi agent extensions, packages, and JSON-RPC SDK clients.

Updated Jul 18, 2026
One-click install
npx skills add https://github.com/arthrod/conejo-skills --skill pi-extending
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pi-extending
Source: https://github.com/arthrod/conejo-skills/tree/main/skills/pi-extending
Command: npx skills add https://github.com/arthrod/conejo-skills --skill pi-extending

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill provides a centralized knowledge base for developers building on the Pi agent ecosystem, resolving ambiguity around extension authoring, package distribution, and programmatic embedding.

Core Features & Use Cases

  • Extension Authoring: Guidance on registering tools, commands, and custom providers using the ExtensionAPI.
  • Package Management: Instructions for distributing skills via npm or git with correct resource precedence.
  • RPC & SDK Integration: Technical specifications for embedding Pi into Node.js/TypeScript applications or using the JSON-RPC interface for bidirectional control.
  • Use Case: A developer needs to create a custom model provider or integrate Pi into a larger TypeScript application; this skill provides the exact API surface and framing requirements to ensure compatibility.

Quick Start

Use the pi-extending skill to explain the required framing and record delimiter rules for implementing a custom JSON-RPC client for the Pi agent.

Frequently Asked Questions about pi-extending

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

FAQPage Schema
How do I build a custom Pi agent extension in TypeScript?

To build a Pi agent extension, use the ExtensionAPI to register custom tools, commands, and providers. This process requires adhering to specific framing, resource resolution, and session management patterns defined in the Pi mono-repo documentation.

How do I implement a JSON-RPC client for the Pi agent?

Implementing a custom JSON-RPC client for the Pi agent requires following specific framing and record delimiter rules. This enables bidirectional control and seamless integration with external systems over the protocol.

Can I embed the Pi agent directly into a Node.js application?

Yes, you can embed the Pi agent directly into Node.js applications using the programmatic SDK client. This allows for in-process agent embedding and custom model provider integration within your TypeScript environment.

What is the correct way to distribute Pi agent packages?

Distributing Pi agent packages involves publishing skills via npm or git while ensuring correct resource precedence. Proper package management guarantees your extensions resolve dependencies correctly across different environments.

What are the limitations when extending Pi agent capabilities?

Limitations when extending Pi agent capabilities revolve around strict framing requirements and session management patterns. Failing to follow the specific resource resolution rules defined in the Pi mono-repo documentation will cause integration failures.