Grove – fast source insights for coding agents – via Tree-sitter
WASM grammar loading without recompilation beats static tree-sitter bindings.

Tree-sitter indexing cuts AI token costs by 15%—actual benchmarks included.
Backend developers, AI engineers building code-aware agents
Sourcegraph · grep · ripgrep
The code indexing tools released in December now have added capabilities (Rust, Perl support), better token efficiency, easier install method, and a website! The website has a wasm port of the query tool, qi, so you can try it out in the browser.
Let me know what you think. Thanks!
WASM grammar loading without recompilation beats static tree-sitter bindings.
Tree-sitter + FTS5 + MCP = tokens saved for AI agents to actually code, not search.
AST + embeddings for codebase search—but Sourcegraph Cody, Cursor, and Continue already solve this.
Tree-sitter + agentic flow ~50x faster than raw LLM calls, but codebase visualization is crowded category.
Tree-sitter AST blueprints compress codebase context when JinaAI and Firecrawl already exist.
AST-aware codebase search and onboarding beats grep, ships as local MCP server.