skills.rest
cy-wali
cy-waliCOMMUNITY
Explore all skills by this author →
GitHub

search-strategy

Transform natural language questions into parallel source-specific searches and synthesize answers.

1|Updated Mar 19, 2026
Search Tags:#multi-source#search#synthesis#deduplication#ranking#query-decomposition#enterprise-search
One-click install
npx skills add https://github.com/cy-wali/knowledge --skill search-strategy-cy-wali
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: search-strategy
Source: https://github.com/cy-wali/knowledge/tree/main/enterprise-search/skills/search-strategy
Command: npx skills add https://github.com/cy-wali/knowledge --skill search-strategy-cy-wali

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The skill transforms a natural language question into parallel, source-specific searches across connected sources, enabling fast, comprehensive insights from multiple knowledge bases.

Core Features & Use Cases

  • Decomposes queries into per-source search strategies (semantic and keyword variants).
  • Executes searches in parallel across all connected sources and merges results.
  • Produces a ranked, deduplicated synthesis suitable for executive summaries and decision-making.
  • Use Case: When you need a quick, cross-source answer from chat logs, docs, and project trackers.

Quick Start

Ask a natural language question and let this skill decompose it into per-source searches and synthesize the results.

Frequently Asked Questions about search-strategy

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

FAQPage Schema
How do I search across multiple knowledge bases and get a single answer?▼

Multi-source search decomposes a natural language question into parallel, source-specific queries across connected knowledge bases, then deduplicates and synthesizes the results into a single coherent answer.

What is query decomposition and how does it improve enterprise search?▼

Query decomposition translates a natural language question into semantic and keyword variants tailored to each connected source, enabling comprehensive parallel searches and higher quality result ranking.

Can I use natural language to search across project trackers and chat logs?▼

Yes, you can ask a natural language question and the skill will execute parallel searches across connected sources like project trackers, chat logs, and docs, then merge the findings into a unified summary.

How to handle search failures when one data source is unavailable?▼

Cross-source search handles unavailable sources through graceful degradation, continuing to execute parallel queries on remaining connected sources and synthesizing answers from available data.

How does multi-source search deduplicate and rank overlapping results?▼

After executing parallel searches, the skill ranks and deduplicates overlapping results before synthesizing them into a single coherent answer, ensuring concise and non-redundant insights for decision-making.

Related Skills

Explore More Data & Analytics Skills→
Supporter09Supporter09

llamaindex

Build retrieval-augmented generation applications over private documents and data sources.

Community
Intermediate
lhohanlhohan

bx

Search the web with token-budgeted results via the bx CLI.

Community
Advanced
ccfccf

hybrid-search-implementation

Combine vector similarity with keyword matching for hybrid retrieval.

Community
Intermediate
ilove323ilove323

search-strategy

Decompose natural language queries into parallel, source-specific searches across connected data sources.

Community
Advanced