search

Runs parallel searches across codebases, documentation, and the web for references and patterns.

5|2|Updated Dec 11, 2025
One-click install
npx skills add https://github.com/thoreinstein/opencode-config --skill search-thoreinstein
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: search
Source: https://github.com/thoreinstein/opencode-config/tree/main/skill/search
Command: npx skills add https://github.com/thoreinstein/opencode-config --skill search-thoreinstein

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill accelerates codebase exploration by running parallel searches across local code, documentation, and the web. It unifies results from code files, docs, and external sources for rapid discovery.

Core Features & Use Cases

  • Parallel Local & External Search: Simultaneously scan code, docs, and web sources for patterns and references.
  • Structured Result Modes: Supports direct matches, related code, and usage patterns to help comprehension.
  • Use Case: Imagine you need to locate every usage of a function across a large repository and its documentation; this Skill returns all relevant matches to act on.

Quick Start

Use the search skill to locate all occurrences of a function named 'parseData' across code, docs, and web.

Frequently Asked Questions about search

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

FAQPage Schema
How do I search for function usages across local codebases and web documentation simultaneously?

To locate function usages across local codebases and web documentation simultaneously, use a parallel multi-source search approach. It runs concurrent scans across local repository files, project docs, and external web sources to discover all relevant references and related code patterns.

What is the best way to discover code patterns across multiple repositories and documentation?

The best way to discover code patterns across multiple repositories and documentation is using layered search techniques. This applies local file discovery via glob, grep, and AST parsing, combined with external research and optional git history checks to reveal related code and usage patterns.

How does parallel codebase search work when scanning local files and external sources?

Parallel codebase search works by running deterministic background tasks across local and external sources concurrently. It unifies structured results from local code files, project documentation, and web sources to accelerate codebase exploration and return direct matches alongside related usage patterns.

Can I use glob and grep to find references in local repositories and project docs at the same time?

Yes, you can use glob and grep to find references in local repositories and project docs concurrently. The search process implements layered local file discovery that runs in parallel with external web research, returning all relevant matches without sequential delays.

Does multi-source code search support git history checks for deeper codebase exploration?

Yes, multi-source code search supports git history checks for deeper codebase exploration. It implements layered search with local file discovery and external research, applying optional deep-dive checks into git history to track references and usages across the repository's evolution.

What are the limitations of parallel search across local code and external web sources?

Parallel search across local code and external web sources relies on deterministic background tasks, meaning it requires exact pattern inputs rather than semantic understanding. It focuses on returning direct matches and related code, so complex contextual reasoning may require additional manual review.