TypeScript Repository Indexing for Code Agent Retrieval

TL;DR

This paper introduces abcoder-ts-parser, a TypeScript parser that creates efficient, reliable code indexes for large repositories, enhancing context retrieval for code agents by leveraging direct compiler API integration.

Contribution

It presents a new TypeScript parser built on the compiler API that overcomes bottlenecks of LSP-based resolution, enabling scalable code indexing.

Findings

The new parser is significantly more efficient than existing architectures.

It produces reliable indexes for large open-source TypeScript projects.

Evaluation on projects up to 1.2 million lines demonstrates scalability.

Abstract

Graph-based code indexing can improve context retrieval for LLM-based code agents by preserving call chains and dependency relationships that keyword search and similarity retrieval often miss. ABCoder is an open-source framework that parses codebases into a function-level code index called UniAST. Its existing parsers combine lightweight AST parsers for syntactic analysis with language servers for semantic resolution, but because LSP-based resolution requires a JSON-RPC call for each symbol lookup, these per-symbol calls become a bottleneck on large TypeScript repositories. We present abcoder-ts-parser, a TypeScript parser built on the TypeScript Compiler API that works directly with the compiler's AST, semantic information, and module resolution logic. We evaluate the parser on three open-source TypeScript projects with up to 1.2 million lines of code and find that it produces reliable indexes significantly more efficiently than the existing architecture. For a live demonstration, watch: https://youtu.be/ryssr7ouvdE