SemaSeek is a Visual Studio Code extension that provides natural-language semantic search for your codebase. Instead of relying on keywords, SemaSeek understands the meaning behind your search queries to find the most relevant code snippets.
Features
Semantic Search: Use natural language to search your code. For example, you can search for "function to calculate fibonacci" instead of just "fibonacci".
Automatic Indexing: SemaSeek automatically indexes your workspace files in the background after manually indexing on the first run (codebase).
File Watching: The index is kept up-to-date by watching for file creations, changes, and deletions.
Persistent Storage: The search index can be persisted to a SQLite database, so it doesn't need to be rebuilt every time you open VS Code. A JSON file fallback is also available.
Commands
Semantic Search: Index Workspace: Indexes the entire workspace.
Semantic Search: Query: Initiates a semantic search query.
Semantic Search: Clear Index: Deletes the search index.
Semantic Search: Show Index Stats: Shows the number of indexed chunks.
Configuration
You can configure SemaSeek through the VS Code settings (settings.json):
semanticSearch.persistToSqlite (default: true): Whether to persist the index to a SQLite database. If false, a JSON file is used as a fallback.
semanticSearch.embeddingModel (default: "gemini-embedding-001"): The embedding model to use. If you change this, you will need to re-index your workspace.
semanticSearch.batchSize (default: 20): The batch size for embedding requests. You can adjust this to manage API rate limits and costs.
Setup
Install Semaseek from the marketplace
Index your workspace:
Open a workspace you want to index.
Open the command palette and run Semantic Search: Index Workspace.
Query your codebase:
Open the command palette and run Semantic Search: Query.
Type the search terms, phrase or keyword in the search bar
