Enables advanced semantic search, SQL querying, and result ranking for efficient repository exploration.
RepoQL Search Patterns provides a specialized interface for navigating complex codebases using a combination of semantic, lexical, and symbol-based search methods. It allows users to perform high-level intent-based exploration with the repoql_explore function or execute precise queries using SQL-style syntax. With built-in support for scoping by path and language, as well as the ability to boost or penalize results using regex, this skill helps developers quickly locate specific implementation patterns, understand architectural flows, and filter out noise like test files or mocks during the discovery phase.
Key Features
01Granular scoping by file path, URI globs, and programming language
020 GitHub stars
03Symbol-specific searching for functions and types with wildcard support
04Advanced result ranking using regex-based boost and penalize parameters
05Intent-based code exploration with ranked results
06Hybrid semantic and lexical search using SQL syntax
Use Cases
01Locating specific symbols or type definitions within a polyglot repository using wildcards
02Filtering search results to focus on core logic while demoting tests, mocks, and specification files
03Mapping out complex architectural flows like authentication or data validation across a large project
