search

Search connected MCP enterprise sources with parallel subqueries and return ranked results.

1|Updated Mar 30, 2026
One-click install
npx skills add https://github.com/ilove323/comlan-skills --skill search-ilove323
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: search
Source: https://github.com/ilove323/comlan-skills/tree/main/enterprise-search/skills/search
Command: npx skills add https://github.com/ilove323/comlan-skills --skill search-ilove323

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables fast, comprehensive enterprise search by querying all connected MCP sources in parallel and synthesizing the results into a single, actionable answer, eliminating manual cross-tool hunting and fragmented context.

Core Features & Use Cases

  • Parallel, source-specific searches across chat, email, cloud storage, project trackers, CRM, and knowledge bases.
  • Intelligent parsing of user intent, entities, time constraints, source hints, and filters (from:, in:, after:, before:, type:).
  • Deduplication, relevance ranking, and result grouping with metadata (timestamps, authors, links) and graceful handling of partial failures.
  • Use case: prepare a meeting brief by aggregating recent decisions mentioned across Slack channels, related emails, and the project tracker, with authoritative documents surfaced first.

Quick Start

Ask the assistant to "Search for recent decisions about the Aurora API redesign across all connected sources and show the most authoritative references with links and dates."

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 enterprise sources like chat, email, and CRM at the same time?

To search across multiple enterprise sources, this Skill queries all connected MCP sources in parallel for a single user query. It applies source-specific subqueries to chat, email, cloud storage, trackers, and CRM, then synthesizes findings into one answer.

Can I filter enterprise search results by sender, date, or source?

Yes, you can filter enterprise search results using specific syntax. The Skill intelligently parses filters like from:, in:, after:, before:, and type: to narrow down results across connected sources based on your constraints.

How does cross-source search handle duplicate content and rank results?

Cross-source search handles duplicates by applying deduplication to the aggregated findings. It then ranks results by relevance and groups them with metadata like timestamps, authors, and links, ensuring authoritative references are surfaced first.

What happens when an enterprise search query fails to retrieve data from one of the connected sources?

When a query fails to retrieve data from a connected source, the Skill provides graceful handling of partial failures. It returns the available synthesized results alongside an error report detailing which sources failed.

How do I prepare a meeting brief by finding recent decisions across different tools?

To prepare a meeting brief, ask the assistant to search for recent decisions across all connected sources. The Skill aggregates relevant findings from chat, email, and trackers, surfacing authoritative documents with links and dates first.

Does semantic search work for finding specific people or project statuses across connected MCP sources?

Yes, semantic search works for finding people and project statuses across connected MCP sources. The Skill parses user intent for decision, document, people, and status queries, returning deduplicated, ranked results with metadata.