r-adb-mcp-server

An MCP server that exposes a large set of tools for controlling Android devices/emulators through ADB, including device/app management, UI inspection and interaction, screen capture/recording, logcat/diagnostics, performance stats, network controls, developer and accessibility toggles, file transfer, and emulator simulation features.

Evaluated Apr 04, 2026 (17d ago)
Homepage ↗ Repo ↗ Automation mcp adb android flutter ui-testing automation quality-assurance visual-qa
⚙ Agent Friendliness
56
/ 100
Can an agent use this?
🔒 Security
16
/ 100
Is it safe for agents?
⚡ Reliability
21
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
55
Documentation
55
Error Messages
0
Auth Simplicity
95
Rate Limits
0

🔒 Security

TLS Enforcement
10
Auth Strength
10
Scope Granularity
0
Dep. Hygiene
40
Secret Handling
30

No authentication/authorization, TLS, or scoped permissions are documented for the MCP server. The toolset includes high-risk actions (install/uninstall apps, reboot, shell_command, file transfer, network/proxy toggling). Because the implementation details are not provided here, dependency and secret-handling hygiene cannot be verified; treat outputs/commands as potentially sensitive and avoid running this in untrusted contexts.

⚡ Reliability

Uptime/SLA
0
Version Stability
35
Breaking Changes
20
Error Recovery
30
AF Security Reliability

Best When

You have an Android device/emulator connected locally and want an MCP-enabled agent to drive ADB-based inspection and testing workflows end-to-end.

Avoid When

You need strong guarantees around command authorization, secrets protection, or remote/public exposure of the MCP server.

Use Cases

  • Automated UI testing and visual QA for Flutter/Android apps
  • Programmatic device management (install/uninstall apps, start/stop, clear data)
  • UI hierarchy inspection and targeted interactions (tap, scroll, input by text/id)
  • Debug workflows using logcat, Flutter logs, crash logs, and ANR traces
  • Visual capture and recording for regression testing and Figma comparisons
  • Device simulation for QA (GPS spoofing, call/SMS simulation)

Not For

  • Production security-critical device management in untrusted environments
  • Use cases requiring authenticated multi-tenant access to remote devices (no such mechanism documented)
  • Environments where running arbitrary shell commands on the host/device is disallowed

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
Yes
SDK
No
Webhooks
No

Authentication

Methods: None documented (local MCP server started via command)
OAuth: No Scopes: No

No authentication/authorization mechanism is described in the provided README/manifest content; interaction appears to be through a locally run MCP server and the connected device's ADB permissions.

Pricing

Free tier: No
Requires CC: No

Open-source (MIT) and installable via PyPI; costs are primarily your infrastructure/agent/runtime and the device/ADB environment.

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • Operations depend on correct ADB connectivity and device state; tools may fail if device is offline, unauthorized, or in an unexpected UI state.
  • UI interactions (tap/scroll) can be sensitive to screen layout changes and element finding accuracy.
  • Recording/start-stop and logcat-like tools can have timing constraints and may require device readiness.
  • Some tools (e.g., shell_command, push/read file, install/uninstall) can have strong side effects; agents should be cautious about invoking them without constraints.

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for r-adb-mcp-server.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-04-04.

8642
Packages Evaluated
17761
Need Evaluation
586
Need Re-evaluation
Community Powered