pane 1 — pitch ↵ read

One call.
not ten reads.

Gortex indexes your repository into an in-memory knowledge graph and serves it to your coding agent over MCP, HTTP, and a web UI. smart_context replaces 5–10 file reads, cutting token usage by ~94%.
Built for 15 AI coding agents — one binary, zero services.

$ curl -fsSL https://get.gortex.dev | sh [copy]
50
mcp tools
92
languages
15
agents wired
~94%
fewer tokens
pane 2 — session · agent@repo live
§01 · whytoken accounting

Stop paying your agent to re-read the same files.

A typical "make an edit" loop in Claude Code or Cursor does half a dozen Reads, a Grep, then maybe a Glob. Gortex answers the same question in one graph query — with callers, callees, imports,
blast radius, and interface implementors attached.

without gortexbaseline
11,480
tokens / edit
  • Read internal/watcher/watch.go (412 lines)
  • Read internal/graph/patch.go (289 lines)
  • Read internal/graph/types.go (603 lines)
  • Grep "OnFileChange" across workspace
  • Read 4 more files to resolve callers
  • Guess at the interface it satisfies
with gortex▶ smart_context
688
tokens / edit
  • 7 symbols · 3 files · signatures + callers
  • interface satisfied: graph.Patcher
  • blast radius: 4 symbols, 2 tests
  • no scan, no guessing, no re-reads
  • live-synced via fsnotify watcher
  • Δ −94.0% vs baseline
§02 · what's in the boxnine pillars

A graph, a server, a daemon, and a hook — in one static binary.

Zero external dependencies. Everything runs in-process, in memory, on your laptop. Sigstore-signed releases, SLSA Level 3 build provenance.

01 knowledge graph core

Every file, symbol, import, call chain, and type relationship in one queryable structure. Tarjan SCC, Louvain communities, blast radius, dead-code detection, hotspot ranking — all in Go.

02 92 languages parse

Tree-sitter + regex extractors. Core (Go, TS, Python, Rust, Java, C#, Kotlin, Swift), templates (Blade, Jinja, ERB), blockchain (Solidity, Move, Cairo), scientific (Julia, R, MATLAB), build/data (SQL, Protobuf, HCL).

03 50 mcp tools agent

Full surface: smart_context, get_edit_plan, verify_change, analyze, scaffold, rename_symbol, edit_file + 7 resources + 3 prompts.

04 http server api

Versioned /v1/* JSON API on :4747 exposing all MCP tools, plus /v1/graph (force-directed dump) and /v1/events (SSE graph-change stream). Bearer-token auth, configurable CORS. Powers IDE plugins, CI hooks, and the Next.js web UI.

05 long-living daemon shared

One process, one graph, every editor window. Live fsnotify on every tracked repo, per-client session isolation, multi-server roster. launchd/systemd auto-start at login — no sudo.

06 gcx1 wire format −27%

Round-trippable compact text instead of JSON, opt-in via format: "gcx" on 13 list-shaped tools. Median −27.4% tiktoken savings (best case −38.3%) across the 20-case bench.

07 watch mode live

Surgical patches on file change across all tracked repos. Your agent's context stays current on every keystroke — no re-index between turns. ~200ms incremental restore from snapshot vs 3-5s full re-index.

08 guards & verification safe

Project-specific rules in .gortex.yaml. verify_change catches signature breaks across all callers and impls. analyze would_create_cycle blocks bad imports before they land.

09 semantic + eval measured

BM25 + vector with RRF fusion. Hugot bundled (pure-Go ONNX, MiniLM-L6-v2). First-class gortex eval covers recall, embedders, GCX1 tokens, SWE-bench. Published baseline: R@1 42.3% · R@5 56.4% · R@20 69.9%.

§03 · architecturecat ARCHITECTURE.txt

Index. Serve. Watch.
In-process.

No database to run. No container to orchestrate. One binary, one graph, two ports — and a Unix-socket daemon when you want to share it across editor windows.

gortex binary ── single static · ~20MB · CGO-on
   
   ├─ CLI (cobra) ───────────────────┐
   ├─ MCP Server (stdio + sse) ──────┤
   ├─ HTTP Server (:4747 /v1/*) ─────┤   ◀── web UI · IDE plugins · CI
   └─ Daemon (~/.gortex/server.sock) ┤   ◀── shared graph for every editor window
                                     ▼
                              ┌──────────────────────────┐
                              │   In-memory graph        │
                              │   nodes · edges · index  │
                              │   on-disk snapshot       │
                              └──────────────▲───────────┘
                                             
   ┌─ Indexer ──► tree-sitter ◀── 92 langs ──┤
   │                                         │
   ├─ Resolver ─► LSP · go/types · SCIP ─────┤   tier-aware: lsp_resolved → ast_inferred → text_matched
   │                                         │
   └─ Watcher ──► fsnotify · debounced ──────┘   per-file patches
§04 · language coverage92 languages · 8 categories

Speaks what your repo speaks.

Structural parsing via tree-sitter. IMPLEMENTS inference for Go, TypeScript, Java, Rust, C#, Scala, Swift, and Protobuf — no extends keyword required.

#
language · deep extraction
fns
methods
types
interfaces
calls
§05 · quickstartthree steps · runs anywhere

Running in three commands.

macOS + Linux · amd64 + arm64. The installer detects your OS, downloads the signed release tarball, verifies SHA256 (and cosign if installed), drops the binary in $HOME/.local/bin. No silent sudo.
Re-runs upgrade in place.

01 · install
$ curl -fsSL https://get.gortex.dev | sh
# or: brew · .deb · .rpm · .apk
# or: go install github.com/zzet/gortex/cmd/gortex@latest

One-time machine setup: gortex install writes user-level MCP wiring + skills + slash commands at ~/.claude/ and ~/.gemini/.

02 · wire your agent
$ cd ~/projects/myapp
$ gortex init --analyze
# writes .mcp.json, CLAUDE.md,
# per-community SKILL.md, hooks

Auto-detects every agent in scope and writes a community-routing block to each — CLAUDE.md, AGENTS.md, .windsurfrules, GEMINI.md, .cursor/rules.

03 · serve + watch
$ gortex daemon start --detach
# or: gortex mcp --watch (per-repo)
# or: gortex server --index .

Daemon runs in the background, supervises the graph for every tracked repo. Auto-start at login: gortex daemon install-service.

§06 · works with15 first-class adapters

Every coding agent
you already run.

Gortex is MCP-native. If your agent speaks the Model Context Protocol, it speaks Gortex. gortex init auto-detects each agent in scope and writes the right config.

Claude Code
slash commands, skills, hooks (PreToolUse, PreCompact, Stop)
● configured
Cursor
.mcp.json + .cursor/rules/gortex-communities.mdc
● configured
Kiro
steering files + agent hooks + .mcp.json
● configured
Windsurf
.windsurfrules + community routing block
● configured
VS Code / Copilot
workspace MCP config + instructions
● configured
Continue.dev
.continue/config.json MCP block
● configured
Cline
cline_mcp_settings.json + AGENTS.md
● configured
OpenCode
opencode.json + AGENTS.md routing
● configured
Antigravity
~/.gemini/antigravity Knowledge Items
● configured
Codex CLI
stdio MCP, scriptable, CI-friendly
● configured
Gemini CLI
~/.gemini settings + GEMINI.md
● configured
Zed
.zed/settings.json MCP wiring
● configured
Aider
.aider.conf.yml + AGENTS.md routing
● configured
Kilo Code
.kilocode/mcp.json auto-discovery
● configured
OpenClaw
openclaw config + community routing
● configured
§07 · get startedexit 0

Teach your agent
what your code knows.

Source-available under a PolyForm-based license — free for individuals, OSS projects, small businesses (<50 staff / <$500K rev), nonprofits, and education. Zero dependencies. One static binary. Sigstore-signed, SLSA-3 build provenance, OpenSSF Scorecard tracked, VirusTotal 0/91.

recent · v0.18.2
· long-living daemon (launchd/systemd)+
· http server /v1/* on :4747+
· gcx1 wire format (−27.4% median)+
· 50 mcp tools, 7 resources, 3 prompts+
· 92 languages parsed+
· 15 agent adapters wired+
· hugot semantic search bundled+
· gortex eval — recall · embedders · swebench+
· web UI at gortexhq/web+
● main
watcher: running
graph: 50 tools · 7 resources · 3 prompts
mcp: stdio · http: :4747 · daemon: socket
v0.18.2
source-available