FridaSAFE
MCP stdio server for frida
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
A Model Context Protocol (MCP) implementation for Frida dynamic instrumentation toolkit.
Overview
This package provides an MCP-compliant server for Frida, enabling AI systems to interact with mobile and desktop applications through Frida's dynamic instrumentation capabilities. It uses the official MCP Python SDK to enable seamless integration with AI applications.
Demo
https://github.com/user-attachments/assets/5dc0e8f5-5011-4cf2-be77-6a77ec960501
Features
- Built with the official MCP Python SDK
- Comprehensive Frida tools exposed through MCP:
- Process management (list, attach, spawn, resume, kill)
- Device management (USB, remote devices)
- Interactive JavaScript REPL with real-time execution
- Script injection with progress tracking
- Process and device monitoring
- Resources for providing Frida data to models
- Prompts for guided Frida analysis workflows
- Progress tracking for long-running operations
- Full support for all MCP transport methods
Installation
Prerequisites
- Python 3.8 or later
- pip package manager
- Frida 16.0.0 or later
Quick Install
pip install frida-mcp
Development Install
# Clone the repository git clone https://github.com/yourusername/frida-mcp.git cd frida-mcp # Install in development mode with extra tools pip install -e ".[dev]"
Claude Desktop Integration
To use Frida MCP with Claude Desktop, you'll need to update your Claude configuration file:
- Locate your Claude Desktop configuration file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json - Linux:
~/.config/Claude/claude_desktop_config.json
- Add the following to your configuration file:
{
"mcpServers": {
"frida": {
"command": "frida-mcp"
}
}
}Usage
Once installed, you can use Frida MCP directly from Claude
81363dda7a52OBSERVED · 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 frida-mcp -- uvx frida-mcp
{
"mcpServers": {
"frida-mcp": {
"command": "uvx",
"args": [
"frida-mcp"
]
}
}
}Exposed tools (6)
5 read · 0 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
enumerate_devices | read | List all devices connected to the system. |
enumerate_processes | read | List all processes running on the system. |
get_device | read | Get a device by its ID. |
get_local_device | read | Get the local device. |
get_session_messages | destructive | Retrieve and clear messages sent by persistent scripts in a session. |
get_usb_device | read | Get the USB device connected to the system. |
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 | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (1 observation(s))
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (2)
get_session_messages
docs/assets/frida-mcp.mov
Gates applied: no_behavioural_pass.
81363dda7a52full audit observations/trust-audit/mcp-server/dnakov__frida.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 | 81363dda7a52 | SAFE | B | 89 | first audit |
Questions
What is the Frida MCP server?
MCP stdio server for frida
What tools does Frida expose?
6 in total: 5 read-only, 0 that write, and 1 that can delete or overwrite (get_session_messages). Every one is listed on this page with its risk.
Is Frida 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. Separately from the audit: 1 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Frida need?
No credential environment variables were found in its source, so it appears to need none.
How current is this page?
The grade is for one exact copy of the source (81363dda7a52), read on 2026-10-01. The repository is watched and re-audited when it changes.