Skip to content
MCP server

MobiLoop

Let an AI agent build, test on a device, and verify Android and iOS apps through Appium

Runs locallyStatus and history
Pricing
Open source
Platforms
CLI, self-hosted, macOS
Runs
Locally
Version
0.1.0-alpha.15
Listing updated
2 days ago

About MobiLoop

MobiLoop is an open-source MCP server for developer tools. It runs locally on your machine and is started by your AI client.

MobiLoop is a controlled layer between a coding agent and a real mobile environment, built around a loop of code change, build, install on a device, Appium test, evidence, classification and report. The README is clear about scope: it provides guarded build-test-verify loops and failure classification, while fully automated patch-and-retest stays outside the default orchestrator until stricter approval and rollback controls exist. This is an alpha release.

Evidence is central. Runs keep build logs, screenshots, Appium page source, logcat or simulator logs, API responses and screenshot diffs together, and logcat evidence is classified into app bugs, automation errors, missing environment, remote rules and test-data issues. Flow memory records screen checkpoints and replays a stable path. A built-in security loop reads Android and iOS configuration through security.scan_source, compares scans before and after a fix and applies a release gate.

Guardrails include server-side approval for high-impact tools, secret path blocking, symlink resolution and redaction of tokens, keys, emails and phone numbers. It runs as a non-root Docker container over stdio, while emulators, devices and Appium stay on the host. It targets Flutter, React Native, Android and iOS projects.

Install

Run in a terminal

claude mcp add mobiloop-mcp -- docker run -i --rm ghcr.io/enessubass/mobiloop-mcp:0.1.0-alpha.15

Add to claude_desktop_config.json

{
  "mcpServers": {
    "mobiloop-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/enessubass/mobiloop-mcp:0.1.0-alpha.15"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "mobiloop-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/enessubass/mobiloop-mcp:0.1.0-alpha.15"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "mobiloop-mcp": {
      "type": "stdio",
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/enessubass/mobiloop-mcp:0.1.0-alpha.15"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "mobiloop-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/enessubass/mobiloop-mcp:0.1.0-alpha.15"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "mobiloop-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/enessubass/mobiloop-mcp:0.1.0-alpha.15"
      ]
    }
  }
}

Configuration

  • MOBILOOP_WORKSPACE_ROOTWorkspace path mounted into the container.
  • APPIUM_SERVER_URLAppium server URL reachable from the container.

Tools this server provides (1)

Taken from the project documentation; the running server may differ.

security.scan_source
Reads Android and iOS configuration, credential, transport, storage and logging signals.

What it does

Evidence-first runs
Build logs, screenshots, page source, device logs and iteration records are stored together.
Appium automation
Semantic taps, typing, swipes, assertions and accessibility summaries on emulators and devices.
Failure classification
Sorts logcat evidence into app bugs, automation errors and environment problems.
Security loop
Scans source and platform settings, compares fixes and gates release.
Approval gate
Secure mode requires approval payloads for high-impact tools.
Flow memory
Records checkpoints and replays the latest passing path.

Good for

  • Have an agent run a Flutter or React Native app on an emulator and report failures
  • Collect screenshots and logs as proof of a fix
  • Scan an Android or iOS project for credential and transport issues
  • Re-run the same path after an approved repair
  • Generate candidate end-to-end scenarios from source

Pricing

Open source

Prices change. Check enessubass.github.io before you buy.

MobiLoop FAQ

Which apps does it target?

Flutter, React Native, Android and iOS apps via Appium.

How do I run it?

As the container image `ghcr.io/enessubass/mobiloop-mcp` over stdio; Appium and devices stay on the host.

Does it automatically fix code?

Not by default. Patch-and-retest is kept outside the default orchestrator pending stricter controls.

Is it stable?

The site labels the current release alpha.15.

Is MobiLoop free?

Yes. MobiLoop is open source, so you can use and self-host it for free.

How do I add MobiLoop to Claude Code?

Run this command in your terminal: claude mcp add mobiloop-mcp -- docker run -i --rm ghcr.io/enessubass/mobiloop-mcp:0.1.0-alpha.15

What are the best alternatives to MobiLoop?

The closest alternatives to MobiLoop are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to MobiLoop

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools