# Research

Studies of coding agents on real C and C++ code: what they get right, what they cost, and how we measured it.

1. **[Why LSP, Why the LSP Plugin, and Why Claude Code Doesn't Save Tokens](https://byteask.ai/research/lsp-claude-code-cpp):** Language Servers, Grep, and the Cost of Code Navigation for C/C++ Coding Agents Nov 4, 2026 Paper · Preprint We ran Claude Code with and without Anthropic's clangd LSP plugin on 255 C and C++ issues from 40 repositories. The configured plugin resolved no more issues and cost 11% more, and on change requests built so that grep returns same-name decoys, plain Claude Code still found 96.6% of the right callers by reading the code. Pratyush Saini, CTO, ByteAsk Research; Anirudha Kulkarni, CEO, ByteAsk [Read the paper](https://byteask.ai/research/lsp-claude-code-cpp) [PDF, 213 KB](https://byteask.ai/papers/why-lsp-claude-code-cpp-2026-11.pdf) [Code](https://github.com/ByteAsk/lsp-claude-code-cpp)

---
HTML version: https://byteask.ai/research