codebase-memory-mcp vs gortex
Code intelligence MCP in pure C: tree-sitter knowledge graph over 158 languages, average repo indexed in milliseconds, sub-ms queries, 10x fewer tokens. Single static binary, zero deps. — versus — Code-intelligence engine in one static Go binary: tree-sitter graph over 257 languages, compiler-grade resolution for 17, multi-repo, 175 configurable MCP tools — up to 50x fewer tokens. 100% local.
The closest fight on the shelf: both are single-binary tree-sitter knowledge graphs served over MCP. codebase-memory-mcp bets on C-core minimalism; gortex on language breadth and cross-service resolution.
| codebase-memory-mcp | gortex | |
|---|---|---|
| Stars | 37k | 1.1k |
| Forks | 2.9k | 98 |
| Language | C | Go |
| License | MIT | Apache-2.0 |
| Last activity | 2 days ago | yesterday |
| Topics | code-intel, local | code-intel, local |
| Curated connections | 6 | 3 |
codebase-memory-mcp — the curator's take
The performance ceiling of the code-context-server category: pure C, single static binary, the Linux kernel indexed in 3 minutes, structural queries under a millisecond — with a peer-reviewed preprint (83% answer quality, 10x fewer tokens across 31 repos) instead of vibes. Hybrid LSP adds real type resolution for the 12 languages that matter most, and 43 client surfaces means it plugs into whatever agent you run. NOT semantic search — it answers structural questions (call chains, routes, blast radius), not 'where's the code that does X'; pair it with an embedding tool for that. And note its own disclosure: it writes to your agent config files by design — audit posture is unusually good (SLSA 3, OpenSSF, per-release VirusTotal), use it.
gortex — the curator's take
The coverage play on this shelf: 257 grammars, cross-repo and cross-SERVICE edges (HTTP routes, contracts) with a provenance/confidence tier the single-repo tools don't attempt, and one install that configures 19 coding agents. When NOT: 175 MCP tools is a context bill of its own — prune the toolset or your agent pays the index's price in schema tokens; benchmarks are self-published; and if you want memory as plain files, this is the opposite philosophy.