grove gives coding agents structural, byte-precise, token-cheap access to a codebase via tree-sitter — instead of reading whole files. One engine, seven tools, two faces (a human CLI grove <verb> and an MCP server grove serve), with grammars loaded at runtime from a WASM registry, so adding a language needs no recompile and no toolchain on the consumer.

(asciinema cast: docs/assets/grove_demo.cast — play it interactively with asciinema play docs/assets/grove_demo.cast.)

Why grove

Agents burn tokens and round-trips grep-ing and read-ing whole files to answer "where is this defined / what does it do / who calls it." grove replaces that with one symbol at a time, by exact bytes, behind a stable id the agent passes between turns.

• Token-cheap — outline a 1700-line file as a skeleton; source one symbol's body, not the whole file. A map call returns a directory's definitions + references in one shot.
• Byte-precise & stable — every result carries a symbol-id (<lang>:<relpath>#<name>@<line>, 1-based) you pass forward across turns.
• One engine, two faces — the same Rust binary drives the CLI and an MCP server, so a human and an agent see the same thing.
• Runtime grammars — all 27 official tree-sitter grammars load from a hosted WASM registry; new languages are a registry entry, not a recompile.

Not an LSP. grove is a syntactic, tree-sitter-powered shell for agents — not a semantic language server. It speaks MCP (not LSP), parses (doesn't analyze), and locates (doesn't refactor): no type inference, completion, rename, or type-resolved go-to-def. It's the cheap syntactic layer beneath where an LSP's semantics begin — complementary, not competitive. Full reasoning: Is grove an LSP?.

27 languages out of the box — one binary, grammars loaded at runtime from the hosted WASM registry:

Bash	C	C++	C#	Go	Java	JavaScript	Julia	PHP	Python
Ruby	Rust	Scala	TypeScript	TSX	Agda2	CSS2	Embedded Template2	Haskell2	HTML2
JSDoc2	JSON2	OCaml2	OCaml Interface2	CodeQL2	Regex2	Verilog2

² minimal profile — core tools only (callers/definition degrade); full profile = all tools. <kbd> = no official logo. Profiles are data, not compiled in. See Languages & grammars.

See VISION.md for the product vision.

Quick start

# 1. install (one line — detects platform, verifies sha256)
curl -fsSL https://raw.githubusercontent.com/Entelligentsia/grove/main/install.sh | sh

# 2. wire it into a project (in the project root)
grove init

grove init detects the project's languages, auto-fetches their grammars, and writes .mcp.json (the tools exist) + a CLAUDE.md steering directive (the agent reaches for grove instead of grep) + grove.lock. That's it — your agent now has structural sight. Other install channels (Homebrew, npm, cargo, agent skill) and --as mcp|skill|both are in Install and Setup.

As an agent skill (Claude Code, Cursor, Codex, Cline, …): npx skills add Entelligentsia/grove — the skill self-installs the binary on first use if it's missing. See Setup.

Evaluated on real codebases

The eval is Entelligentsia/grove-testbench: it runs the same agent (Claude) on the same prompt with grove off (baseline) and grove on (grove) across 10 large, popular, grammar-backed codebases, and measures the impact on context tokens, wall-clock time, turns, and answer quality — same agent both sides, grove the only variable. It's evidence-first (blind-judged answers verified against pinned source), not a highlight reel; where grove regresses, it's reported and filed as a fix.

Early numbers — L2 callsites, run 2, grove v0.1.7 (1 run/side; full data in the testbench's FINDINGS.md):

• Context tokens: median −93% (range −89% to −97% on the winners) — e.g. redis 2.81M → 83K (−97%), rails 1.98M → 128K (−94%), webpack 4.52M → 307K (−93%), django 1.40M → 95K (−93%), tokio 1.22M → 98K (−92%), hugo 4.52M → 377K (−92%), laravel 3.76M → 432K (−89%).
• Tool calls: median −91% (range −77% to −96%) — fewer, sharper hops instead of grepping and whole-file reads.
• Wall-clock: median −64% (range −48% to −82%) where comparable.

Honest caveat: grove is not a universal win yet — on spring-boot it regressed (+36% context, slower) for the L2 callsites task; that's tracked in the testbench's GROVE-ISSUES.md. The other two charted repos in the testbench are TypeScript and bitcoin (tool calls −96% / −59%; their baseline context was delegated to subagents so isn't on the token chart). The full per-repo breakdown, methodology, and raw runs: Entelligentsia/grove-testbench.

The tools

Add --json to any command for the agent-facing shape. Full reference + examples: Tools.

Documentation

• Install — curl/Homebrew/npm/cargo, build from source, the agent skill
• Setup — grove init, --as mcp|skill|both, what it writes, offline/dry-run
• Languages & grammars — the WASM registry, fetch/lock, where grammars live, profiles
• Tools — the seven tools, --json, symbol-id, examples
• MCP server — grove serve, .mcp.json, steering, error model
• Roadmap & repo layout — what's not done yet, source map
• FAQ — Is grove an LSP? and other positioning questions
• VISION.md — product vision · CHANGELOG.md — releases
• Eval: Entelligentsia/grove-testbench — same agent, grove off vs on, across 10 large repos
• Registry: Entelligentsia/grove-registry · Homebrew tap: Entelligentsia/homebrew-grove

Status

Pre-1.0. callers/definition are name-based (no receiver-type resolution); 12 languages ship a minimal profile (core tools only); no incremental reparse yet. Details and the rest of the roadmap: Roadmap.