code-review-graph vs serena
Two of the top projects in MCP servers for developers, side by side: who leads on each score part and who each one suits, then setup, license, activity and what each review found.
code-review-graph
Local Tree-sitter code graph that gives coding assistants review context over MCP
serena
MCP server giving coding agents symbol-level code search, editing and refactoring
Verdict
code-review-graph ranks higher, 92 to 91. Of the 4 score parts we measure for both, code-review-graph leads on 2, serena on none and 2 are even.
- Adoptioncode-review-graph leads by 1 point, 91 to 90: close to even.
- FreshnessEven at 100.
- Maintenancecode-review-graph leads by 3 points, 93 to 90: close to even.
- Agent-readyEven at 70.
Pick code-review-graph if you are
Developers who want coding assistants to read fewer files when reviewing changes
- One install command detects and configures about 16 AI coding tools, with a clean uninstall
- Incremental updates re-parse only SHA-256-changed files; two-file edit on django took about 2.5 s
Pick serena if you are
Developers who want their coding agent to navigate and refactor large codebases
- Symbol-level retrieval and editing instead of line numbers or text search
- Language server backend covers over 40 languages and is free
| What we compare | code-review-graph | serena |
|---|---|---|
| Score parts, out of 100 | ||
| Adoption | 91, widely used | 90, widely used |
| Freshness | 100, active | 100, active |
| Maintenance | 93, healthy | 90, healthy |
| Agent-ready | 70, partly | 70, partly |
| Facts from GitHub and the README | ||
| Stars | 32.7k | 31.3k |
| License | MIT (permissive) | custom license (read the license) |
| Last commit | Nov 2026 | Nov 2026 |
| Last release | Nov 2026 | Oct 2026 |
| Language | Python | Python |
| Docker | Yes | Yes |
| GPU | Not needed | Not stated |
| arm64 or Apple Silicon | Not stated | Not stated |
code-review-graph
Parses a repository with Tree-sitter into a SQLite graph of functions, classes, imports, calls and tests, then serves blast-radius and risk-scored context over MCP so the assistant reads only affected files. The installer configures about 16 tools (Claude Code, Cursor, Codex, Windsurf, Zed and others) and updates incrementally via hooks, git hooks or watch mode. A GitHub Action posts risk-scored PR comments.
Who it is for: Developers who want coding assistants to read fewer files when reviewing changes
Strengths
- One install command detects and configures about 16 AI coding tools, with a clean uninstall
- Incremental updates re-parse only SHA-256-changed files; two-file edit on django took about 2.5 s
- Covers 30+ languages plus notebooks, and custom grammars via languages.toml
- Runs locally; the GitHub Action builds the graph on the CI runner without sending code out
Weaknesses
- Impact recall of 1.0 is circular, since ground truth comes from the same graph
- Against a grep baseline, median token reduction is 6.65x, not the 63x headline
- Graph context can cost more tokens than a plain read for small single-file edits
- Cold build of a roughly 3,000-file repository took about 40 seconds
- no GPU
- Docker
- Needs Python 3.10+, Tree-sitter, SQLite
- Models: all-MiniLM-L6-v2 (local embeddings)
serena
Serena is an MCP server that exposes IDE-style tools to a coding agent: find symbol, find referencing symbols, symbol overview, rename, and replace or insert at symbol level. It uses language servers (LSP) for over 40 languages by default, or a paid JetBrains plugin that adds move, inline, type hierarchy and a debugger. It plugs into MCP clients such as Claude Code, Codex, Gemini-CLI, VSCode and Claude Desktop, and includes a per-project memory system.
Who it is for: Developers who want their coding agent to navigate and refactor large codebases
Strengths
- Symbol-level retrieval and editing instead of line numbers or text search
- Language server backend covers over 40 languages and is free
- Works with any MCP client; stdio launch command or HTTP mode
- Layered YAML config: global, per-project, per-client context, and modes
Weaknesses
- Move, inline, type hierarchy and debugging need the paid JetBrains plugin
- Language server backend only renames symbols; no move or inline
- Some languages need extra language server dependencies installed manually
- Mixed licensing: GPL-3.0-or-later app with MIT SolidLSP component, plus a CLA
- Docker + Compose
- Needs uv, Python 3.13, language servers per language (LSP backend), JetBrains IDE and Serena plugin (optional, paid)
- Models: Any LLM through an MCP client
More in MCP servers for developers
Building with these?
We help teams pick an open-source AI stack and run it in production. Client work never changes a rank.