Can this skill distinguish between a casual chat and an official document?

Yes, it uses an authority hierarchy that prioritizes official wikis for factual policy questions while prioritizing chat history for decision-making context.

How does the Search Strategy skill handle different data sources?

It translates natural language into specific syntaxes for different enterprise tools, using semantic search for conceptual queries and keyword search for specific entities or project names.

What happens if a search returns no results?

The skill employs automated fallback strategies, such as broadening keywords, removing date constraints, or suggesting query modifications to the user.

Does it perform searches sequentially or in parallel?

It utilizes parallel execution to search all connected sources simultaneously, ensuring the total search time is only as long as the slowest single source.

Enterprise Search Strategy

Name: Enterprise Search Strategy
Author: anthropics

byanthropics

•

8,000

•

Productivity & Workflow

Orchestrates multi-source enterprise searches by decomposing natural language queries into targeted, source-specific sub-queries.

The Search Strategy skill acts as the intelligent orchestration layer for enterprise knowledge retrieval, transforming complex natural language questions into parallelized searches across chat platforms, wikis, and project trackers. By analyzing user intent and classifying query types—such as decisions, status updates, or official documentation—it applies specialized ranking logic and fallback strategies to ensure the most relevant and authoritative information is surfaced from fragmented data sources. It bridges the gap between conversational AI and siloed corporate data by translating intent into actionable, source-specific syntax.

Key Features

01Intelligent query decomposition into source-specific sub-queries

02Parallel search execution for low-latency cross-platform retrieval

038,000 GitHub stars

04Automated filter mapping for chat, wiki, and project management tools

05Intent classification for prioritized source selection (Decision, Status, Factual)

06Authority-based result ranking and deduplication logic

Use Cases

01Retrieving specific technical specs or company policies from enterprise wikis

02Aggregating project status updates from task trackers and activity logs

03Locating project decisions buried in fragmented chat threads and meeting notes

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add anthropics/knowledge-work-plugins search-strategy

For use in Claude.ai and ChatGPT

Download Skill