topos-mcp 0.4.1

Topos MCP server — structural code-quality tools (SIMPLE/COMPOSABLE/SECURE) for AI coding agents.

A structural quality gate for coding agents.

Topos measures complexity, coupling, and risky data flows, then gives your agent a concrete target—from SLOP to GOLD.

Install · What it checks · Under the hood · Docs · Issues

---

Why Topos

Coding agents produce working code quickly. The harder question is whether the result is still easy to understand, safe to change, and well-fitted to the rest of the repository. Quality is the new currency.

Topos computes that signal from program structure—not from an LLM review or a style opinion—and returns concrete failure locations and next actions. It is fast enough to sit inside the agent loop: measure, edit, verify, repeat.

Tests check behavior. Topos checks whether the implementation is built to keep changing.

Grounded in category theory, powered by a native-Rust engine.

Under the hood

Topos is a self-contained Rust CLI and MCP server. Analysis runs locally; your source code is not sent to an external model or hosted analysis service.

Component	Role
tree-sitter	Parses six languages and powers the native AST, CFG, CPG, PDG, and UAST representations.
GitNexus	Supplies the repository dependency graph scored by COMPOSABLE (topos depgraph generate). Requires npm install -g gitnexus@1.6.8.
Sighthound	Embedded in the MCP server for supplementary security findings; native CPG probes remain the SECURE scoring source.
Graphify	Optional advisory orphan and fragile-edge detection via topos graphify / topos_refactor(target="graphify"); does not affect the medal. Requires pip install graphifyy.

The result is one agent-facing contract over several structural lenses: one score to optimize, explicit evidence for each failure, and a verification loop that can tell a real improvement from cosmetic churn.

Install and Quick Start

VS Code MCP Extension

Open the Extensions view, search @mcp topos, select Topos, and choose Install. Or view here: Topos: GitHub's MCP Registry.

Then ask agent mode:

Use Topos to find this repository's worst structural problem, make one focused improvement, and verify the result.

See the agent setup guide for tool permissions and troubleshooting.

Other MCP clients [Claude Code]

Run the self-contained MCP server on demand—no persistent Topos or Python installation required:

claude mcp add --transport stdio topos -- uvx topos-mcp

Setup for Codex, Gemini CLI, Cursor, Windsurf, Antigravity, and manual JSON lives in the agent setup guide.

Standalone CLI

curl -fsSL https://docs.krv.ai/topos/install.sh | sh

Prefer Homebrew?

brew install krv-labs/tap/topos

On Homebrew 6+, that one-liner auto-taps and trusts only this formula. If you brew tap krv-labs/tap first, run brew trust --formula krv-labs/tap/topos before brew install topos.

Then run:

topos evaluate . -r

[!TIP] Want GOLD? Run topos depgraph generate if your agent hasn't already (+ run again after big structural edits), then evaluate with the added CLI flag --gitnexus-dir .gitnexus.

Topos supports Python, Rust, JavaScript, TypeScript, C++, and Go. The CLI defaults to Python; use --language rust|go|javascript|typescript|cpp for another language. See Installation for platform support and alternative install paths.

More ways to use Topos

• OpenClaw / ClawHub: openclaw skills install @Krv-Labs/topos
• Hermes: hermes skills tap add Krv-Labs/topos then hermes skills install Krv-Labs/topos/topos
• MCP Registry name: io.github.Krv-Labs/topos
• CLI reference: docs.krv.ai/topos/cli

What Topos checks

Every file gets three independent verdicts:

• SIMPLE — avoids unnecessary complexity using AST entropy and control-flow complexity.
• COMPOSABLE — stays decoupled from the repository using module-dependency structure and Martin instability.
• SECURE — avoids dangerous API reachability and taint paths in the code property graph.

Those verdicts roll up into one memorable quality medal without hiding which pillar failed:

Medal	Criteria
🥇 GOLD	Passes all 3
🥈 SILVER	Passes 2 of 3
🥉 BRONZE	Passes 1 of 3
❌ SLOP	Passes 0, or fails to parse

Topos also returns ranked refactor guidance: failing metric locations, control-flow cycles, load-bearing dependency edges, process bottlenecks, and optional Graphify knowledge-graph findings. Advisory findings never silently change the scored medal.

How the medal system is derived

The three pillars are pairwise incomparable and form an eight-element evaluation lattice; GOLD is their intersection.

---
config:
  layout: dagre
  theme: neutral
---
graph BT
    SLOP["❌ SLOP<br/>No Medal"]
    SIMPLE["🥉 BRONZE<br/>Simple"]
    COMPOSABLE["🥉 BRONZE<br/>Composable"]
    SECURE["🥉 BRONZE<br/>Secure"]
    SC["🥈 SILVER<br/>S ∧ C"]
    SSc["🥈 SILVER<br/>S ∧ Sc"]
    CSc["🥈 SILVER<br/>C ∧ Sc"]
    IDEAL["🥇 GOLD<br/>Quality Code"]

    SLOP --> SIMPLE
    SLOP --> COMPOSABLE
    SLOP --> SECURE
    SIMPLE --> SC
    SIMPLE --> SSc
    COMPOSABLE --> SC
    COMPOSABLE --> CSc
    SECURE --> SSc
    SECURE --> CSc
    SC --> IDEAL
    SSc --> IDEAL
    CSc --> IDEAL

    style SLOP       fill:#f8d7da,stroke:#842029,color:#000
    style SIMPLE     fill:#cd7f32,stroke:#5c3a1e,color:#fff
    style COMPOSABLE fill:#cd7f32,stroke:#5c3a1e,color:#fff
    style SECURE     fill:#cd7f32,stroke:#5c3a1e,color:#fff
    style SC         fill:#c0c0c0,stroke:#4a4a4a,color:#000
    style SSc        fill:#c0c0c0,stroke:#4a4a4a,color:#000
    style CSc        fill:#c0c0c0,stroke:#4a4a4a,color:#000
    style IDEAL      fill:#ffd700,stroke:#856404,color:#000

Measures · Category-theory foundations

Distribution

Topos ships four ways:

• GitHub Releases — the topos CLI binary (macOS/Linux), via install.sh or a direct release download.
• PyPI — topos-mcp, a thin bin-wheel bundling the MCP server binary (pip install topos-mcp / uvx topos-mcp), zero Python runtime.
• VS Code Marketplace — the Topos extension, bundling platform binaries.
• Docker — a container image for Glama and other MCP-registry hosting.

Crate layout and adapter details: docs.krv.ai/topos/architecture.

Contributing

Topos is used internally at Krv Labs to manage AI-agent code output. We welcome bugs, ideas, and contributions.

• Bug? Open an issue
• Idea? Start a discussion or open a PR
• Collaborate? [e-mail]

---

Full documentation · Measures and metrics · Engineering notes