search

Search repository code by semantic meaning using the mcp__mira__code tool.

8|2|Updated Jul 18, 2025
One-click install
npx skills add https://github.com/ConaryLabs/Mira --skill search-conarylabs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: search
Source: https://github.com/ConaryLabs/Mira/tree/main/plugin/skills/search
Command: npx skills add https://github.com/ConaryLabs/Mira --skill search-conarylabs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The semantic code search skill helps developers locate relevant code by meaning and intent rather than by exact text, reducing time spent on manual code hunting.

Core Features & Use Cases

  • Semantic code search: retrieve implementations by concept (authentication, error handling, API usage) across large codebases.
  • Result-rich outputs: show file path, line numbers, relevance score, and a snippet preview to quickly validate relevance.
  • Use Case: quickly find the function or pattern across modules when you remember the behavior but not the exact symbol or string.

Quick Start

Use the skill to search for "authentication middleware" or "database connection pooling" across the repository.

Frequently Asked Questions about search

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

FAQPage Schema
How do I find code by meaning instead of exact text matches?

Semantic code search locates implementations by concept and intent rather than exact text. It retrieves relevant code across large codebases by analyzing the meaning of your query to surface matching functions and patterns.

What is the best way to search for authentication patterns across a repository?

The best way to find authentication patterns across a repository is semantic code search. It retrieves implementations by concept, returning file paths, line numbers, relevance scores, and code previews to validate relevance.

How does semantic code search handle large codebases?

Semantic code search handles large codebases by applying meaning-based retrieval to locate relevant implementations. It returns up to 10 results by default, each including file paths, line numbers, relevance scores, and snippet previews.

Can I search for API usage implementations without knowing the exact symbol name?

Yes, you can search for API usage implementations without knowing the exact symbol name. Semantic code search locates code by behavior and intent, returning relevant implementations with file paths and line numbers for quick validation.

What details are included in semantic code search results?

Semantic code search results include file paths, line numbers, relevance scores, and code snippet previews. These details allow developers to quickly validate the relevance of the retrieved implementations without opening the files.

Does semantic code search work for finding error handling logic across modules?

Yes, semantic code search works for finding error handling logic across modules. It applies meaning-based retrieval to locate relevant implementations by concept, returning matched code with file paths and relevance scores for validation.