search

Search the operation's knowledge base and return top-ranked abstracts.

Updated Apr 26, 2026
One-click install
npx skills add https://github.com/kinnerkarmanish/mak --skill search-kinnerkarmanish
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: search
Source: https://github.com/kinnerkarmanish/mak/tree/main/templates/enterprise/skills/search
Command: npx skills add https://github.com/kinnerkarmanish/mak --skill search-kinnerkarmanish

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Quickly locate relevant information across references, documents, and signals to support decision-making and reduce manual digging.

Core Features & Use Cases

  • Full-text search across reference files, domain knowledge, and signals to surface top-ranked abstracts.
  • Reuse results in agent workflows to inform planning, debugging, and client interactions.
  • Use Case: an agent searches for pricing strategy docs to inform a deployment plan.

Quick Start

Type a query like /search "pricing strategy" to retrieve the most relevant results.

Frequently Asked Questions about search

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

FAQPage Schema
How do I perform full-text search across a knowledge base for relevant information?

Full-text search parses your query to retrieve top-ranked abstracts from references, domain knowledge, and signals, enabling agents to discover information for research and problem solving.

Can I use full-text search to find specific documents for agent workflows?

Yes, full-text search surfaces top-ranked abstracts from reference files and domain knowledge, which agents can reuse in workflows to inform planning, debugging, and client interactions.

What is the best way to search references and signals to support decision-making?

The best way to search references and signals is querying the knowledge base with full-text search, which applies relevance ranking to quickly locate relevant information and reduce manual digging.

How does relevance ranking work when retrieving abstracts from a knowledge base?

Relevance ranking evaluates parsed queries against the knowledge base to return top results as L0 abstracts, ensuring the most pertinent information from references and signals is surfaced first.

When do I need full-text search for my agent's research and planning tasks?

You need full-text search when agents must discover information across references, domain knowledge, and signals to inform research, planning, and problem solving without manual document digging.

Are there limitations to searching a knowledge base using abstracts?

Search returns top results as L0 abstracts rather than full documents, meaning agents receive summarized relevance snippets which may require further retrieval for complete context.