ck:scout

Scout codebases to locate relevant files and surface concise file summaries.

Updated Apr 2, 2026

Search Tags:#codebase #search #gemini #task-management #opencode #parallel-agents #scout

Ready to run

npx skills add https://github.com/hidonguyen/trade-ops --skill ck-scout-hidonguyen

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: ck:scout
Source: https://github.com/hidonguyen/trade-ops/tree/main/.opencode/skills/scout
Command: npx skills add https://github.com/hidonguyen/trade-ops --skill ck-scout-hidonguyen

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Rapidly discover and summarize relevant files across large codebases so engineers and reviewers can locate context and surface next actions without manual searching.

Core Features & Use Cases

Parallel agent scouting: Splits the repository into segments and runs concurrent subagents to maximize coverage and reduce wall-clock search time.
Internal and external modes: Supports lightweight internal Explore agents or external Gemini/OpenCode CLI agents depending on scale and available tools.
Task registration and observability: Creates and updates scout tasks with metadata, enforces 3-minute timeouts, and aggregates results with deduplication and unresolved questions.
Use cases include initial feature discovery across multiple directories, preparing for large refactors, locating migration or auth files for debugging, and onboarding contributors.

Quick Start

Use the scout skill to find authentication-related files across the repository and return a concise list of file paths with one-line descriptions.

Frequently Asked Questions about ck:scout

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

FAQPage Schema

How do I search a large multi-directory codebase to find relevant files quickly?▼

Codebase scouting uses parallel agent spawning to split repositories into segments and run concurrent subagents, maximizing coverage and reducing wall-clock search time for locating relevant files.

What is the best way to surface concise file summaries for feature discovery and onboarding?▼

Parallel codebase scouting discovers and summarizes relevant files across large repositories, returning concise file paths with one-line descriptions to help engineers locate context without manual searching.

Can I use external Gemini or OpenCode CLI agents for codebase scouting?▼

Yes, codebase scouting supports internal lightweight Explore agents or external Gemini and OpenCode CLI agents, allowing you to choose depending on scale and available tools.

How do I manage context window limits when reading and summarizing large codebases?▼

Codebase scouting enforces 3-minute timeouts and uses chunked file reading to respect context window limits, aggregating results with deduplication to manage large repositories efficiently.

Does parallel codebase scouting support task registration and observability?▼

Codebase scouting creates and updates scout tasks with metadata, enforcing 3-minute timeouts while aggregating results with deduplication and unresolved questions for full observability.