shared-knowledge

Search and index a shared company knowledge base using the qmd CLI.

Updated Apr 28, 2026
One-click install
npx skills add https://github.com/khoabd/paperclip --skill shared-knowledge
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: shared-knowledge
Source: https://github.com/khoabd/paperclip/tree/main/skills/shared-knowledge
Command: npx skills add https://github.com/khoabd/paperclip --skill shared-knowledge

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates the wasted effort and inconsistent decisions that occur when AI agents duplicate research, miss existing architecture decisions, and lack access to cross-team company context.

Core Features & Use Cases

  • Hybrid Knowledge Search: Perform keyword, semantic, or hybrid searches across the shared company knowledge base to find architecture decisions, agent specs, and existing research.
  • Knowledge Contribution: Add new findings, research results, and formal Architecture Decision Records (ADRs) to the shared knowledge base for all agents to access.
  • Use Case: If an agent is tasked with building a new AI adapter, it can first search the shared KB to see if another agent already researched compatible adapters or made related architecture decisions, avoiding redundant work.

Quick Start

Use the shared-knowledge skill to search the company knowledge base for existing research on AI adapter architecture before starting new integration work.

Frequently Asked Questions about shared-knowledge

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

FAQPage Schema
How do I prevent AI agents from duplicating research and missing existing architecture decisions?

To prevent duplicating research and missing architecture decisions, use a shared knowledge base that enables AI agents to perform hybrid semantic and keyword searches for existing cross-team context before starting new work.

How do I search existing architecture decision records across cross-team company context?

You search existing architecture decision records by executing keyword or semantic searches across the shared company knowledge base using the qmd CLI, retrieving relevant agent specs and reusable research.

Can I add new findings and architecture decision records to a shared knowledge base for other agents?

Yes, you can add new findings, research results, and architecture decision records to the shared knowledge base. The qmd CLI indexes new content to maintain accessibility for all cross-team agents.

Does shared-knowledge work with the qmd CLI for hybrid semantic search?

Yes, shared-knowledge relies on the qmd CLI to execute hybrid semantic and keyword searches, index new content, and maintain the shared knowledge base at the default Paperclip instance shared directory.

What is the best way to share reusable research and agent specs across multiple AI agents?

The best way to share reusable research and agent specs is maintaining a shared knowledge base where agents contribute findings and query existing context via hybrid search, eliminating inconsistent cross-team architecture decisions.