ck:scout

Coordinate parallel agents to locate files and map project structure.

Updated Mar 30, 2026
One-click install
npx skills add https://github.com/mafvietnam/mafapp --skill ck-scout-mafvietnam
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ck:scout
Source: https://github.com/mafvietnam/mafapp/tree/main/.claude/skills/scout
Command: npx skills add https://github.com/mafvietnam/mafapp --skill ck-scout-mafvietnam

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Fast, token-efficient codebase scouting using parallel agents to find files needed for tasks.

Core Features & Use Cases

  • Internal Explore: Scans code directories in parallel to map file locations and relationships.
  • External Scout: Optionally leverages Gemini/OpenCode for broader searches when needed.
  • Use Case: Rapidly understand project structure for onboarding, debugging, or feature scoping across large repos.

Quick Start

Analyze the user prompt to identify search targets and spawn Explore subagents to scan the codebase in parallel, then aggregate the results.

Frequently Asked Questions about ck:scout

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

FAQPage Schema
How do I explore a large codebase to understand its structure quickly?

To explore a large codebase quickly, you can use parallel agents to scan directories like src/, lib/, and tests/ simultaneously. This coordinates multiple searches to map file locations and relationships for rapid onboarding or debugging.

What is the best way to search for specific files when debugging across a large repository?

The best way to search for files during debugging is using parallel scouting agents that scan assigned scopes simultaneously. This approach locates relevant files and aggregates results efficiently, reducing overall search time.

How does parallel codebase scouting work for feature scoping?

Parallel codebase scouting works by spawning Explore subagents that scan different directories simultaneously. A Task-based orchestration protocol assigns specific scopes to each agent, enforces a 3-minute timeout, and aggregates the findings.

Can I use external tools like Gemini or OpenCode for broader codebase searches?

Yes, you can optionally leverage external tools like Gemini or OpenCode for broader codebase searches. The system uses internal Explore subagents by default but supports integrating these external tools when expanded search capacity is needed.

Does parallel codebase scouting work without external dependencies?

Yes, parallel codebase scouting works without external dependencies by using internal Explore subagents by default. It coordinates these built-in agents to scan directories and map project structure without requiring any additional tools.

What are the limitations of using parallel agents for codebase discovery?

A key limitation is a strict 3-minute timeout enforced per agent during codebase discovery. If an agent cannot complete its directory scan within this window, its specific scope results may be incomplete or omitted from the final aggregation.