DISCOVER THE FUTURE OF AI AGENTS

sentrux

Added Apr 24, 2026
Agent & Tooling
Open Source
Workflow AutomationRustModel Context ProtocolAI AgentsCLIAgent & ToolingDeveloper Tools & Coding

Real-time architectural sensor that provides AI coding agents with a feedback loop on code architecture quality through five graph-theoretic dimensions and MCP integration.

sentrux is a real-time architectural quality sensor written in Rust, designed for AI coding agents (Claude Code, Cursor, Windsurf, etc.). It analyzes project file dependencies and computes a non-gameable 0–10000 composite score (Quality Signal) based on five graph-theoretic dimensions—modularity, acyclicity, depth, equality, and redundancy—visualized through a real-time interactive Treemap.

Core capabilities include: an MCP Server with 9 built-in tools for AI agent feedback loop integration; session_start/session_end mechanisms to track per-session quality changes; a rules engine via .sentrux/rules.toml supporting constraints on cyclic dependencies, coupling levels, cyclomatic complexity, layering, and boundaries; sentrux check/gate commands for CI/CD quality gates. It supports 52 languages through a tree-sitter plugin system where the binary contains no language-specific code—custom language plugins require only plugin.toml and tags.scm. The GUI renders via wgpu with auto-adaptive Vulkan/GL backends on Linux. The project is MIT-licensed with a Pro tier using Ed25519 key verification.

Installation#

# macOS
brew install sentrux/tap/sentrux

# Linux
curl -fsSL https://raw.githubusercontent.com/sentrux/sentrux/main/install.sh | sh

# Windows
curl -L -o sentrux.exe https://github.com/sentrux/sentrux/releases/latest/download/sentrux-windows-x86_64.exe

# Build from source
git clone https://github.com/sentrux/sentrux.git
cd sentrux && cargo build --release

Quick Start#

  1. Run sentrux /path/to/project to open the GUI and view the Treemap and Quality Signal score.
  2. Create .sentrux/rules.toml in the project root to define architecture constraints.
  3. Run sentrux check . to verify rule compliance.
  4. Run sentrux gate --save . to save a quality baseline.
  5. Configure sentrux --mcp in your MCP client to enable AI agent feedback loop integration.

MCP Configuration (Generic)#

{
  "mcpServers": {
    "sentrux": {
      "command": "sentrux",
      "args": ["--mcp"]
    }
  }
}

Rules Engine Example (.sentrux/rules.toml)#

[constraints]
max_cycles = 0
max_coupling = "B"
max_cc = 25
no_god_files = true

[[layers]]
name = "core"
paths = ["src/core/*"]
order = 0

[[layers]]
name = "app"
paths = ["src/app/*"]
order = 2

[[boundaries]]
from = "src/app/*"
to = "src/core/internal/*"
reason = "App must not depend on core internals"

Unconfirmed Information#

  • Pro pricing details not publicly disclosed
  • Team/company background unclear (4 contributors)
  • Claimed 36K lines of Rust code not independently verified
  • Specific MCP protocol version number not stated

Related Projects

View All

STAY UPDATED

Get the latest AI tools and trends delivered straight to your inbox. No spam, just intelligence.