DroidMindSAFE
Control your Android devices with AI using Model Context Protocol
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://www.python.org/downloads/) [](LICENSE) [](docs/plan.md) [](https://github.com/astral-sh/ruff) [](https://github.com/microsoft/pyright) [](https://modelcontextprotocol.io/) [](https://www.android.com/) [](https://hyperb1iss.github.io/droidmind/)
Control Android devices with AI through the Model Context Protocol
DroidMind is a powerful bridge between AI assistants and Android devices, enabling control, debugging, and system analysis through natural language. By implementing the Model Context Protocol (MCP), DroidMind allows AI models to directly interact with Android devices via ADB in a secure, structured way. When used as part of an agentic coding workflow, DroidMind can enable your assistant to build and debug with your device directly in the loop.
💫 Core Features
DroidMind empowers AI assistants to:
- 📱 Manage Devices: Connect via USB/TCP-IP, list devices, view properties, and r
58a40edf59b0OBSERVED · 2026-10-01Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add droidmind -- uvx droidmind
{
"mcpServers": {
"droidmind": {
"command": "uvx",
"args": [
"droidmind"
]
}
}
}Exposed tools (8)
8 read · 0 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
android-app | read | |
android-device | read | |
android-diag | read | |
android-file | read | |
android-log | read | |
android-screenshot | read | |
android-shell | read | |
android-ui | read |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (2 observation(s))
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (1)
- **Rooted Devices**: If your device is rooted and ADB is configured to run as root, DroidMind will inherit these elevated privileges. This significantly increases the potential impact of any command.
Gates applied: no_behavioural_pass.
58a40edf59b0full audit observations/trust-audit/mcp-server/hyperb1iss__droidmind.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-01 | 58a40edf59b0 | SAFE | B | 89 | first audit |
Questions
What is the DroidMind MCP server?
Control your Android devices with AI using Model Context Protocol
What tools does DroidMind expose?
8 in total: 8 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is DroidMind safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does DroidMind need?
No credential environment variables were found in its source, so it appears to need none.
How does DroidMind run?
It speaks sse and stdio, so it runs as a local process your client starts. It is published on PyPI as droidmind.
How current is this page?
The grade is for one exact copy of the source (58a40edf59b0), read on 2026-10-01. The repository is watched and re-audited when it changes.