zzet/gortex

High-performance code-intelligence engine for AI agents and IDE, supports 257 languages, multi repositories, based on graph, with access via CLI, MCP Server, and API. AI coding agents teammate - expose only needed information, cutting token usage up to 50x. 100% local. Discord: https://discord.gg/39MFHu3J5d

Summary

Gortex is a Go-based code-intelligence engine that indexes repositories into a persistent knowledge graph and exposes it through a CLI, MCP server, HTTP API, and web UI. It gives AI coding agents and IDEs targeted context—such as symbols, call chains, HTTP routes, and cross-repository contracts—without requiring them to read entire files first.

What it is useful for

Gortex is most useful when an engineer or coding agent needs precise code context for a large or multi-repository codebase. Instead of relying only on text search, agents can query relationships in the indexed graph.

Documented use cases include:

• Tracing callers and callees around a function or class.
• Checking the likely impact of changing a symbol.
• Matching API providers and consumers across repositories, including HTTP routes, gRPC services, GraphQL schemas, message topics, WebSocket events, environment variables, OpenAPI paths, and Temporal workflows.
• Previewing an edit or simulating a call chain without changing files.
• Reviewing pull requests with per-PR impact analysis, reviewer suggestions, and line-anchored findings.
• Inspecting the graph visually in the web UI or through versioned /v1/* API endpoints.

The README reports up to 50× fewer tokens per response compared with naive file reads, but that is not guaranteed for every project. Engineers can run gortex savings to inspect measured savings for their own calls and sessions.

Language coverage is broad—257 languages or grammars—but parsing quality is not uniform. The README describes multiple parsing tiers and compiler-grade resolution for a specific set of languages, including Go, Python, JavaScript/TypeScript, Rust, Java, C/C++, and others. The language documentation should be checked before relying on deep resolution for a particular technology.

How engineers can use it

The documented installation uses one script and installs a single binary for macOS, Linux, or Windows:

curl -fsSL https://get.gortex.dev | sh

On Windows PowerShell:

irm https://get.gortex.dev/install.ps1 | iex

Then perform the documented setup:

gortex install
gortex daemon start --detach
gortex track ~/projects/myapp
cd ~/projects/myapp && gortex init

gortex install configures detected coding assistants with MCP access, skills, and slash commands. gortex track adds a repository, while gortex init creates repository-level configuration such as .mcp.json, hooks, and community routing. After that, connected assistants can use graph queries instead of loading broad file contents.

For direct CLI workflows, the README documents commands such as gortex prs, gortex review, and gortex savings. Telemetry is off by default and can be checked with gortex telemetry status.

Building from source requires Go 1.26+ and CGO:

make build
make test
make lint

The README points to docs/onboarding.md for a longer walkthrough and docs/cli.md for the complete command reference.