Searching

Not sure which kit feature to reach for? Use this page as a mental map of search-and-discovery tools – from plain-text grep all the way to LLM-powered semantic retrieval.

Tip: You can mix-and-match different search approaches. For instance, run a docstring search first, then feed the matching files into ContextAssembler for an LLM chat.

Approaches in detail

1. Plain-text / regex search

Fast, zero-setup, works everywhere. Use when you know what string you’re looking for.

repo.search_text("parse_jwt", file_pattern="*.py")

2. AST-based pattern search

Find code by its structure using Abstract Syntax Tree patterns. Perfect for finding specific code constructs like async functions, error handlers, or class inheritance patterns.

from kit.ast_search import ASTSearcher
searcher = ASTSearcher(repo)
# Find all async functions
results = searcher.search_pattern("async def", mode="simple")

3. Symbol indexing

extract_symbols() uses tree-sitter queries (Python, JS, Go, etc.) to list functions, classes, variables – handy for nav trees or refactoring tools.

4. LLM summarization

Generate natural-language summaries for files, classes, or functions with Summarizer. Great for onboarding or API docs.

5. Vector search (raw code)

VectorSearcher chunks code (symbols or lines) → embeds chunks → stores them in a local vector database. Good when wording of the query is similar to the code.

6. Docstring vector search

DocstringIndexer first summarizes code, then embeds the summary. The resulting vectors capture intent, not syntax; queries like “retry back-off logic” match even if the code uses exponential delays without those words.

---

Still unsure? Start with text-search (cheap), move to vector search (smart), and layer summaries when you need meaning over matching.