search

Decompose user queries and run parallel searches across connected MCP sources.

704|58|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/openyak/desktop --skill search-openyak
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: search
Source: https://github.com/openyak/desktop/tree/main/backend/app/data/plugins/enterprise-search/skills/search
Command: npx skills add https://github.com/openyak/desktop --skill search-openyak

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Consolidate queries across all MCP sources into a single, coherent search and return a synthesized answer, reducing time spent cross-referencing tools.

Core Features & Use Cases

  • Parallel multi-source searches: simultaneously query chat, email, cloud storage, project trackers, CRM, and knowledge bases.
  • Query decomposition & synthesis: breaks down user questions, runs parallel searches, and presents a unified result.
  • Use Case: quickly locate a decision, document, or discussion that spans multiple tools (e.g., a decision in a chat thread, a related email, and a file in storage).

Quick Start

Ask to search across all connected MCP sources in one query.

Frequently Asked Questions about search

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

FAQPage Schema
How do I search across multiple MCP sources like email, cloud storage, and project trackers at the same time?

You can search across all connected MCP sources simultaneously by submitting a single query. The search decomposes your question and runs parallel queries across chat, email, storage, CRM, and project trackers, returning a synthesized, ranked results set with source metadata.

What is the best way to find a decision that spans a chat thread, an email, and a file in cloud storage?

Finding a decision that spans a chat thread, an email, and a file requires querying multiple tools at once. This multi-source search breaks down your query, runs parallel searches across connected sources, and presents a unified result consolidating the related discussion and file.

Can I query connected knowledge bases and CRM records together in one search?

Yes, you can query connected knowledge bases and CRM records together in one search. The process identifies all available sources, parses your query, and simultaneously searches across them to produce a single synthesized answer with source metadata.

Do I need to specify which sources to include when running a multi-source knowledge search?

No, you do not need to manually specify sources for a multi-source knowledge search. The process automatically identifies all available connected MCP sources, decomposes your query, and runs parallel searches across them to return a unified, ranked results set.

Does multi-source search work with any connected chat and cloud storage tools?

Multi-source search works with any chat, email, cloud storage, project trackers, CRM, or knowledge bases connected via MCP. It parses your query and runs parallel searches across all identified connections to synthesize a unified result.

Why does my multi-source query return a synthesized answer instead of a list of links?

A multi-source query returns a synthesized answer to reduce time spent cross-referencing tools. It decomposes your question, runs parallel searches across connected sources, and synthesizes the findings into a single, coherent, ranked results set with source metadata.