MikhailHal/ariadne

šŸ’» Developer Tools
0 Views
0 Installs

ā˜• šŸ  šŸŽ 🪟 🐧 - Affected-test selection for Kotlin/Android projects: returns only the unit tests impacted by the current working-tree changes, so agents skip full test runs while iterating. Function-level static analysis via Kotlin Analysis API — call graph over functions, properties (initializers, delegates, accessors), constructors and callable references — with automatic multi-module and Android variant source-set discovery. Verified on Now in Android (34 modules, 4s per query). brew install mikhailhal/tap/ariadne

Quick Install

One-Click IDE Configuration
claude_desktop_config.json
{
  "mcpServers": {
    "mikhailhal-ariadne": {
      "command": "npx",
      "args": [
        "-y",
        "mikhailhal-ariadne"
      ]
    }
  }
}
Or

Using an AI coding agent (Claude Code, Cursor, etc.)? Copy a ready-made prompt that tells it to fetch the setup instructions and install this server for you.

Documentation Overview

ariadne banner

ariadne

MCP Server for Affected Test Selection

Release MCP Registry Homebrew Container image License Kotlin


ariadne is an MCP (Model Context Protocol) server that provides AI agents with the ability to identify affected tests. Powered by sazanami, it analyzes code changes and returns only the tests that need to be run.

ariadne MCP server

[!IMPORTANT]

āš ļø MUST READ: Speed over Completeness

ariadne is built for the agent inner loop — edit, verify, commit — where fast feedback matters more than exhaustive selection. Static analysis cannot trace every execution path: reflection, DI frameworks, and data-flow indirection (e.g., Flux/MVI dispatch) can hide dependencies from any affected-test-selection tool, not just ariadne.

Always keep a final line of defense in CI. Run the full test suite (or a conservative selection) before merging. ariadne narrows what an agent runs while iterating; it is not a replacement for CI.

When ariadne detects changes it cannot analyze (build scripts, resources, unscanned source sets), it says so explicitly in the tool response instead of silently reporting "no affected tests".

Features

  • MCP Integration — Works with Claude Code, Claude Desktop, and other MCP-compatible clients
  • Automatic Git Diff — No need to pass diff manually; ariadne runs git diff internally
  • Powered by sazanami — Uses Kotlin Analysis API for accurate static analysis

Installation

Homebrew (recommended)

brew install mikhailhal/tap/ariadne

Then register it with your MCP client — for Claude Code:

claude mcp add ariadne -- ariadne

Or add to your MCP client configuration manually (e.g., Claude Desktop):

{
  "mcpServers": {
    "ariadne": {
      "command": "ariadne"
    }
  }
}

Docker

docker run -i --rm -v /path/to/project:/workspace ghcr.io/mikhailhal/ariadne

Mount the project you want analyzed and pass /workspace as project_path. The image is also listed in the official MCP Registry as io.github.MikhailHal/ariadne.

Manual (release JAR)

Download ariadne-<version>-all.jar from Releases (requires JDK 21+) and configure your client with "command": "java", "args": ["-jar", "/path/to/ariadne-<version>-all.jar"].

Build from Source

git clone --recursive https://github.com/MikhailHal/ariadne.git
cd ariadne
./gradlew shadowJar   # fat JAR: build/libs/ariadne-<version>-all.jar

Usage

Once configured, AI agents can use the get_affected_tests tool:

Tool: get_affected_tests

Parameters:

  • project_path (required) — Path to the Kotlin project
  • scope (optional, deep | shallow, default deep) — how far back to look:
    • deep — all changes since base_branch (committed and uncommitted). Safest; the whole branch is covered so nothing you already committed slips through unverified. May select more tests.
    • shallow — uncommitted changes only (diff against HEAD). Fastest, for the tight edit loop. Verifying already-committed work is left to the caller. base_branch is ignored.
  • base_branch (optional, deep scope only) — Branch to compare against. When omitted, ariadne uses the repository's default branch (origin/HEAD). If that is not set (e.g. a repo with no remote), it returns an error asking you to pass base_branch explicitly rather than guessing.

Returns:

  • List of affected test FQNs (fully qualified names), sorted
  • If the diff contains changes outside the analyzed Kotlin sources (build scripts, resources, unscanned source sets), a note is appended recommending a full test run for those changes
  • Analysis is bounded by a 120s timeout; on timeout an explicit error is returned

Example

Agent request:

{
  "name": "get_affected_tests",
  "arguments": {
    "project_path": "/path/to/your/kotlin/project"
  }
}

Response:

com.example.UserServiceTest.testCreateUser
com.example.UserRepositoryTest.testSave

How It Works

ā”Œā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”     ā”Œā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”     ā”Œā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”
│  MCP Client │ ──▶ │   ariadne   │ ──▶ │  sazanami   │
│  (Agent)    │     │ (MCP Server)│     │  (Analysis) │
ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜     ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜     ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜
                           │
                           ā–¼
                    ā”Œā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”
                    │  git diff   │
                    ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜
  1. Agent calls tool — Passes project path to ariadne
  2. Run git diff — ariadne executes git diff --unified=0 against base branch
  3. Analyze with sazanami — Build call graph and find affected tests
  4. Return results — List of test FQNs returned to agent

Real-World Validation: Now in Android

Measured against Now in Android (Google's reference Android app — 34 modules, ~268 Kotlin files):

MetricResult
Recall audit — 19 target functions across all layers18/18 valid targets detected (the 19th had no exercising unit test; correctly not selected)
End-to-end response time~4s (module discovery + call-graph build + BFS)
Module discovery34 modules via settings.gradle.kts, incl. nested modules and type-safe accessor dependencies
Source setsmain, debug, prod, benchmark, testDemo, … discovered per module (androidTest* excluded by design)

Verified patterns include repositories behind project interfaces, a library-interface override (androidx.datastore.Serializer), operator fun invoke use cases, @Composable functions, extension mappers, ViewModel property-initializer chains, and callable references. Two representative results:

  • Changing core:common's asResult() selects 14 tests across three modules, including ViewModel tests reachable only through val uiState = ...stateIn(...)
  • Changing the mapper PopulatedNewsResource.asExternalModel() selects 14 tests, including 11 repository tests reachable only through .map(Type::mapper) chains

Test-class selection rate

Every unit-test class in Now in Android was measured by changing a function in the class it tests and checking whether that test class was selected:

Test styleSelected
Plain unit tests (construct the object, call it)13 / 13 valid targets
Robolectric / Compose screenshot tests12 / 12
Framework-dispatched callbacks (lint Detector)0 / 2 — see below

Robolectric turned out not to be a barrier: those tests call the composable themselves (setContent { NiaTheme { ... } }), so the call exists in the source. What decides coverage is not the test runner but whether the test's own code contains the call.

What ariadne cannot see

The rule of thumb: if the framework calls your code instead of your test calling it, ariadne cannot connect them. These are limits of static analysis, not bugs — plan your CI safety net around them:

PatternStatus
Framework-invoked callbacks — Fragment/Activity lifecycle (launchFragmentInContainer), lint Detector methods, Application.onCreateNot traced: no call written in the test
Reflection / DI-container wiringNot traced
UDF dispatch (Flux/MVI)dispatch → collect is never an edge, but wiring in init (or a start() the test calls) is covered conservatively via constructor chains. Subscriptions started by DI/lifecycle are not covered (sazanami#38)
stateIn / shareIn chains (map, onEach, flatMapLatest, combine)Covered — verified with exact selection
Instrumented tests (androidTest*)Out of scope by design
Build scripts, resources, unscanned source setsNot analyzed — reported explicitly in the tool response
Same-name top-level extensions in one packageOver-selected (receiver types are not part of top-level FQNs) — safe direction
KMP source sets (commonMain, expect/actual)Enumerated, but resolution quality unverified (#1)

Full audit notes: sazanami#29, sazanami#38.

Requirements

  • JDK 21 or later
  • Git — For diff detection

Limitations

  • Module discovery is convention-based: it parses settings.gradle(.kts) includes, enumerates src/<sourceSet>/{kotlin,java} layouts, and reads project(":x") / type-safe accessor dependencies from build files. Dynamic includes, projectDir remapping, custom srcDirs, and dependencies injected by convention plugins are not detected — see #1
  • Full graph rebuild on each request (no caching yet); analysis is capped at 120s
  • See What ariadne cannot see for analysis-level gaps

License

Copyright 2025 ariadne contributors

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

See LICENSE for the full text.

Related MCP Servers

Moxie-Docs-MCPā˜… Featured

MCP & Agent Skills for Automated Documentation, and codebase conventions + context

šŸ’» Developer Tools2 views
3KniGHtcZ/codebeamer-mcp

šŸ“‡ ā˜ļø šŸŽ 🪟 🐧 - Codebeamer ALM integration for managing work items, trackers, and projects. Provides 17 tools for reading and writing items, associations, references, comments, and risk management data via Codebeamer REST API v3.

šŸ’» Developer Tools1 views
21st-dev/Magic-MCP

Create crafted UI components inspired by the best 21st.dev design engineers.

šŸ’» Developer Tools0 views
a-25/ios-mcp-code-quality-server

šŸ“‡ šŸ  šŸŽ - iOS code quality analysis and test automation server. Provides comprehensive Xcode test execution, SwiftLint integration, and detailed failure analysis. Operates in both CLI and MCP server modes for direct developer usage and AI assistant integration.

šŸ’» Developer Tools0 views

Engagement

Views
0
Installs
0
Upvotes
0

Views and upvotes are unique per visitor network (hashed IP). Installs count copy actions.

Status

Health: Not checked yet

We have not completed a health check for this listing yet.

No check timestamp yet.

Unclaimed listing (imported or pending owner verification). Claim it →
ā˜… Spotlight Slot

Feature Your MCP Server

Get maximum visibility for your server across our directory, search results, and detail pages.

Spotlight Your Server

Own this project?

This directory is pre-filled from public sources. Claim via GitHub README, site badge, or DNS TXT to get the verified badge and attach your website.

Claim this listing

Promote this listing

Optional paid placement. Free listings stay free forever.

Share & Embed

Add our SVG badge (dark/light directory styles) or embeddable widget to your site.