Halborn Logo

Chromium Browser Extension - zkPass


Prepared by:

Halborn Logo

HALBORN

Last Updated 10/11/2024

Date of Engagement by: August 27th, 2024 - September 10th, 2024

Summary

89% of all REPORTED Findings have been addressed

All findings

9

Critical

0

High

0

Medium

6

Low

3

Informational

0


1. Introduction

zkPass engaged Halborn to conduct a security assessment on their Transgate extension, beginning on 2024-08-27 and ending on 2024-09-10. The security assessment was scoped to the assets provided to the Halborn team.

2. Assessment Summary

The team at Halborn was provided two weeks for the engagement and assigned a full-time security engineer to verify the security of all the applications. The security engineer is a penetration testing expert with advanced knowledge in web, recon, discovery & infrastructure penetration testing and blockchain protocols.

- Improve the security of the implementation

- Identify potential security issues that could be affecting the implementation

During the security assessment of the extension, several vulnerabilities were identified, posing significant risks to the application’s security posture. However, these vulnerabilities have been mostly remediated through specific actions taken, such as patching, code refactoring, or security best practices from the zkPass team.

Key issues included improper input validation, excessive logging of sensitive data, and the use of outdated cryptographic practices. Input handling functions were found to inadequately sanitize user inputs, which could allow attackers to exploit injection vulnerabilities, leading to data compromise or unauthorized access. Sensitive information, including requests and responses, was logged without redaction, exposing the extension to potential data leaks.

Additionally, the use of outdated dependencies presents a major security concern, as these packages may contain known vulnerabilities that have been publicly disclosed. Furthermore, the application relies on TLS 1.2 for secure communication, which, while still supported, lacks the security enhancements provided by TLS 1.3, such as forward secrecy and reduced handshake latency. The absence of encryption for stored session data and the presence of hardcoded sensitive information further exacerbate the risks.

Addressing these vulnerabilities through package updates, implementation of secure coding practices, and adopting modern security protocols will significantly improve the extension's overall security, reducing the likelihood of exploitation and ensuring better protection of user data.

The browser extension is designed with stringent encryption standards, ensuring that all functions and transmitted data are securely protected. No risks to client data privacy were identified in our assessment, confirming that the extension maintains robust safeguards without compromising client information.


3. RISK METHODOLOGY

Vulnerabilities or issues observed by Halborn are ranked based on the risk assessment methodology by measuring the LIKELIHOOD of a security incident and the IMPACT should an incident occur. This framework works for communicating the characteristics and impacts of technology vulnerabilities. The quantitative model ensures repeatable and accurate measurement while enabling users to see the underlying vulnerability characteristics that were used to generate the Risk scores. For every vulnerability, a risk level will be calculated on a scale of 5 to 1 with 5 being the highest likelihood or impact.
RISK SCALE - LIKELIHOOD
  • 5 - Almost certain an incident will occur.
  • 4 - High probability of an incident occurring.
  • 3 - Potential of a security incident in the long term.
  • 2 - Low probability of an incident occurring.
  • 1 - Very unlikely issue will cause an incident.
RISK SCALE - IMPACT
  • 5 - May cause devastating and unrecoverable impact or loss.
  • 4 - May cause a significant level of impact or loss.
  • 3 - May cause a partial impact or loss to many.
  • 2 - May cause temporary impact or loss.
  • 1 - May cause minimal or un-noticeable impact.
The risk level is then calculated using a sum of these two values, creating a value of 10 to 1 with 10 being the highest level of security risk.
Critical
High
Medium
Low
Informational
  • 10 - CRITICAL
  • 9 - 8 - HIGH
  • 7 - 6 - MEDIUM
  • 5 - 4 - LOW
  • 3 - 1 - VERY LOW AND INFORMATIONAL

4. SCOPE

Files and Repository
(a) Repository: Transgate-extension
(b) Assessed Commit ID:
(c) Items in scope:
  • Transgate-extension-main/TransGate/asset-manifest.json
  • Transgate-extension-main/TransGate/ba20a8bed0ccb80b0fe4.wasm
  • Transgate-extension-main/TransGate/index.html
↓ Expand ↓
Out-of-Scope:
Remediation Commit ID:
  • ffdd2ba
Out-of-Scope: New features/implementations after the remediation commit IDs.

5. Assessment Summary & Findings Overview

Critical

0

High

0

Medium

6

Low

3

Informational

0

Impact x Likelihood

HAL-05

HAL-07

HAL-01

HAL-08

HAL-06

HAL-02

HAL-04

HAL-03

HAL-09

Security analysisRisk levelRemediation Date
Improper Input Handling and ParsingMediumSolved - 10/02/2024
Excessive Logging of Sensitive DataMediumSolved - 10/02/2024
Potential Weakness in Key ManagementMediumSolved - 10/02/2024
Lack of Encryption for Stored Session DataMediumRisk Accepted - 10/02/2024
Lack of Validation in Event ListenersMediumSolved - 10/02/2024
Lack of Validation When Parsing JSONMediumRisk Accepted - 10/02/2024
Unvalidated URL ManipulationLowSolved - 10/02/2024
Outdated packagesLowRisk Accepted - 10/02/2024
Use of TLS1.2LowFuture Release - 10/02/2024

6. Findings & Tech Details

6.1 Improper Input Handling and Parsing

// Medium

Description

Improper input handling occurs when user-supplied data is not properly validated, sanitized, or escaped before being processed. This can lead to injection attacks, data corruption, or application behavior manipulation.

In the apiHelper.ts file, the function pickQueryParams parsed user input directly without validation or sanitization. This made it vulnerable to URL parameter manipulation and injection attacks.

Proof of Concept
function pickQueryParams(paramsStr: string) {
  const params: any[] = []
  if (paramsStr) {
    paramsStr.split("&").forEach((str) => {
      const entry = str.split("=")
      params.push(entry)
    })
  }
  return params
}
Score
Impact: 4
Likelihood: 3
Recommendation

Implement strict input validation and sanitation on all user-supplied data. Avoid directly splitting user input and always sanitize before processing.

Remediation

SOLVED: The zkPass team added signature verification for the schema information.

Remediation Hash
ffdd2baf089e7835465a21ea74026eace88721af

6.2 Excessive Logging of Sensitive Data

// Medium

Description

Logging sensitive information such as passwords, tokens, or request details can lead to information leakage, especially if logs are exposed to unauthorized users or external systems.

In httpSession.ts, sensitive data such as request objects were logged without redaction, increasing the risk of data exposure.

Proof of Concept
export default async function TPT(props: TPTProps) {
  const { nodeInfo, requests, condition, clientMsgEncKey, nodeMsgDecKey } = props

  const allResponse: any[] = []
  let attrIndexArr: number[] = []
  const publicFields: any[] = []
  const result: ResponseVerifyInfo[] = []
  const privateDataList: any = []
  let i = 0
  while (i < requests.length) {
    console.log('******************START SEND REQUEST*************************')
    const request = requests[i]
    const api = condition.APIs[i]
    const requestInfo = buildRequestInfo(api, request.header, request.body, allResponse)
    const tls = new TlsClient(requestInfo.host)
    tls.setServerNodeInfo(nodeInfo.host, nodeInfo.tid, clientMsgEncKey, nodeMsgDecKey)
    const res: any = await tls.send(requestInfo.text, requestInfo.accept)
    console.log('tls response info', res)
    checkResponseStatus(res.responseStatus)
    if (res.accept?.indexOf('application/json') > -1 && api.assert) {
      const { indexArr, assert } = formatAssertAndCheckValue(res.json, {
        assert: api.assert,
        HRCondition: condition.HRCondition,
      })
      api.assert = assert
      attrIndexArr = attrIndexArr.concat(indexArr)
    }
Score
Impact: 3
Likelihood: 3
Recommendation

Avoid logging sensitive data, or ensure that logs are securely stored, and sensitive information is redacted or encrypted before logging.

Remediation

SOLVED: The zkPass team reduced significantly the number of output logs on the console, keeping some of them until they have a final stable version of the code.

Remediation Hash
ffdd2baf089e7835465a21ea74026eace88721af

6.3 Potential Weakness in Key Management

// Medium

Description

Poor key management, such as failing to validate key lengths or integrity, can lead to insecure cryptographic operations, exposing the application to brute-force attacks or misuse of weak keys.

In salsa.js, cryptographic keys were used without validation of their length or integrity, increasing the risk of weak key usage.

Proof of Concept
export function Salsa20(key, data){
  // sigma is Salsa's constant "expand 32-byte k"
  const sigma = new Uint8Array([ 101, 120, 112, 97, 110, 100, 32, 51, 50, 45, 98, 121, 116, 101, 32, 107 ])
  const out = new Uint8Array(16)
  core_salsa20(out, data, key, sigma)
  return out
Score
Impact: 3
Likelihood: 3
Recommendation

Implement key length validation and ensure proper key management practices. Use libraries that handle key validation and provide secure key exchange mechanisms.

Remediation

SOLVED: The zkPass team removed the vulnerable code from the code.

Remediation Hash
ffdd2baf089e7835465a21ea74026eace88721af

6.4 Lack of Encryption for Stored Session Data

// Medium

Description

Storing sensitive information, such as session data, without encryption can expose it to unauthorized access if the storage mechanism is compromised.

In storage.ts, session data was stored in chrome.storage.session without encryption, which could expose user session data to potential theft or modification.

Proof of Concept
export async function setSessionStorage(keyMap: any) {
  return await chrome.storage.session.set(keyMap)
}
Score
Impact: 3
Likelihood: 3
Recommendation

Always encrypt sensitive data before storing it, even in browser storage, to prevent unauthorized access.

Remediation

RISK ACCEPTED: The zkPass team stated: "As for the issue mentioned in 6.4, we carefully reviewed it and found that we are only saving the program’s running state and intermediate processes, without involving any user privacy data (the source data is public), so we did not perform encryption."

6.5 Lack of Validation in Event Listeners

// Medium

Description

Failing to validate or sanitize data before handling it in event listeners can expose applications to malicious inputs or unwanted behavior through injection attacks.

In listeners.ts, incoming messages were handled without validation, leaving the application vulnerable to malicious messages that could manipulate internal states.

Proof of Concept
chrome.runtime.onMessage.addListener(async (request) => {
  if (!request) {
    return
  }
  dispatchEvent(request)
  return true
})
Score
Impact: 4
Likelihood: 2
Recommendation

Implement strict validation and sanitation for all incoming event messages, especially those originating from external or user-controlled sources.

Remediation

SOLVED: The zkPass team solved the issue by adding a function to sanitize the request object before performing any operation on the object.

Remediation Hash
ffdd2baf089e7835465a21ea74026eace88721af

6.6 Lack of Validation When Parsing JSON

// Medium

Description

Parsing JSON without validation can lead to the processing of malicious or malformed data, which may result in injection attacks or unexpected behavior.

In jsonParse.ts, JSON data from responses were parsed without validation, making the application vulnerable to manipulation through malicious JSON inputs.

Proof of Concept
function genNodeParams(response: TlsResponse, api: API) {
  console.log('-----genNodeParams-----')
  const { records, fullResponse, responseSliceList, requestInfo, accept } = response

  const resStruct = genDataStruct(responseSliceList, api, accept)

  console.log('-----response struct-----', resStruct)
  console.log('-----records-----', records)
  console.log('-----record length-----', records.length)
Score
Impact: 4
Likelihood: 2
Recommendation

Implement validation and sanitization checks when parsing JSON data, especially when working with untrusted input or external sources.

Remediation

RISK ACCEPTED: The zkPass team stated: "Since our JSON properties and structure are not fixed, we only validated the JSON’s legality."

6.7 Unvalidated URL Manipulation

// Low

Description

Manipulating URLs without proper validation or sanitation can lead to security issues such as URL injection, redirection attacks, or request forgery.

In request.ts, URLs were split and manipulated without validation, which could be exploited for URL injection or improper URL handling.

Proof of Concept
export function buildRequestInfo(api: API, header: any, body: any, dependResponse: any[]): RequestInfo {
  let { url, requestHeaders } = header
  const { method } = header
  const { requestBody } = body
  const verifyFieldPosition: any = {}
  console.log("dependResponse-------", dependResponse)
  console.log("header url-------", url)
  const host = api.host || url.split("/").slice(2, 3)[0]
  url = '/' + url.split("/").slice(3).join("/")
  const end = url.indexOf("?") > -1 ? url.indexOf("?") : url.length

  let query = pickQueryParams(url.substring(end + 1))

  url = url.substring(0, end)
Score
Impact: 3
Likelihood: 2
Recommendation

Ensure proper validation and sanitization of URLs before performing operations on them to prevent injection or manipulation attacks.

Remediation

SOLVED: The zkPass team added a signature verification during the execution, and added as an extra layer of security URL's validation.

Remediation Hash
ffdd2baf089e7835465a21ea74026eace88721af

6.8 Outdated packages

// Low

Description

Multiple outdated packages were identified in the extension, which could significantly increase the risk of security vulnerabilities. These outdated dependencies may contain known security flaws that have been publicly disclosed, leaving the extension vulnerable to a variety of potential attacks. Exploiting these vulnerabilities could lead to issues such as unauthorized access, data breaches, or even complete compromise of the system. By continuing to use outdated packages, the extension may miss out on critical security patches and performance improvements provided by more recent versions, thereby increasing the overall attack surface of the application.

Proof of Concept

Issue

Package                       

Patched version

protobufjs Prototype Pollution vulnerability 

protobufjs

>=7.2.5

PostCSS line return parsing error                            

 postcss                                                      

>=8.4.31

 @adobe/css-tools Regular Expression Denial of Service        

 @adobe/css-tools                                             

>=4.3.1

 @adobe/css-tools Improper Input Validation and Inefficient   

 @adobe/css-tools                                             

>=4.3.2

 Follow Redirects improperly handles URLs in the url.parse()  

 follow-redirects                                             

>=1.15.4

 follow-redirects' Proxy-Authorization header kept across     

 follow-redirects                                             

>=1.15.6

 browserify-sign upper bound check issue in dsaVerify leads 

 browserify-sign                                              

>=4.2.2

 Path traversal in webpack-dev-middleware                     

 webpack-dev-middleware                                       

>=5.3.4

 Express.js Open Redirect in malformed URLs                   

 express                                                      

>=4.19.2

 express vulnerable to XSS via response.redirect()            

 express                                                      

>=4.20.0

 Babel vulnerable to arbitrary code execution when compiling  

 @babel/traverse                                              

>=7.23.2

 Axios Cross-Site Request Forgery Vulnerability               

 axios                                                        

>=1.6.0

 Server-Side Request Forgery in axios                         

 axios                                                        

>=1.7.4

 Uncontrolled resource consumption in braces                  

 braces                                                       

>=3.0.3

 ejs lacks certain pollution protection                       

 ejs                                                          

>=3.1.10

 ws affected by a DoS when handling a request with many HTTP  

 ws                                                           

>=8.17.1

 Regular Expression Denial of Service (ReDoS) in micromatch   

 micromatch                                                   

>=4.0.8

 Webpack's AutoPublicPathRuntimeModule has a DOM Clobbering   

 webpack                                                      

>=5.94.0

 body-parser vulnerable to denial of service when url         

 body-parser                                                  

>=1.20.3

 send vulnerable to template injection that can lead to XSS   

 send                                                         

>=0.19.0

 serve-static vulnerable to template injection that can lead  

 serve-static                                                 

>=1.16.0

 path-to-regexp outputs backtracking regular expressions      

 path-to-regexp                                               

>=0.1.10

 Regular Expression Denial of Service in content              

 content                                                      

>=3.0.7

 Inefficient Regular Expression Complexity in nth-check       

 nth-check                                                    

>=2.0.1

 ws affected by a DoS when handling a request with many HTTP  

 ws                                                           

>=7.5.10

Elliptic allows BER-encoded signatures    

 elliptic                                                     

>=6.5.7

Elliptic's ECDSA missing check for whether leading bit of r and s is zero

 elliptic                                                     

>=6.5.7

Elliptic's EDDSA missing signature length check

 elliptic                                                     

>=6.5.7

Score
Impact: 2
Likelihood: 2
Recommendation

Ensure that all dependencies are updated to the latest stable versions. Use your package manager to check for available updates and apply them. Regular updates would ensure that your extension benefits from the latest security patches and performance improvements.

Remediation

RISK ACCEPTED: The zkPass team stated: "During the upgrade process, we discovered a large number of package incompatibilities, so we have decided not to proceed with the upgrade for now."

6.9 Use of TLS1.2

// Low

Description

TLS 1.2 is still secure but lacks some security enhancements and optimizations of TLS 1.3, including better encryption methods, forward secrecy, and reduced latency.

Proof of Concept
export default class TlsClient {
  constructor(host, tls_version = TLS_VERSION.TLS_1_2) {
    this.handshake_messages = new Buffer()
    this.tls_version = tls_version
    this.host = host
    this.params = new HandshakeParameters()
    this.records = []
    this.appData = ''
    this.appDataLen = -3
    this.acceptType = ''
Score
Impact: 3
Likelihood: 1
Recommendation

Upgrade to TLS 1.3 to benefit from stronger encryption, forward secrecy, and reduced handshake latency. If TLS 1.2 must be used, ensure that only secure ciphers are enabled (e.g., AES-GCM or ChaCha20-Poly1305) and that weaker algorithms (like RSA key exchange) are disabled.

Remediation

PENDING: The zkPass team stated: "For the support of TLS 1.3 mentioned in 6.9, it is included in our future development plans and will not be handled in the current version."

Halborn strongly recommends conducting a follow-up assessment of the project either within six months or immediately following any material changes to the codebase, whichever comes first. This approach is crucial for maintaining the project’s integrity and addressing potential vulnerabilities introduced by code modifications.

© Halborn 2024. All rights reserved.