Back to skill

Security audit

Auth0 Expo

Security checks for vulnerabilities and agentic risk

Overview

The skill is coherent for Auth0 Expo setup, but unsafe token-handling examples and mutable install commands need review before use.

Install only after reviewing the bootstrap steps and active Auth0 tenant. Do not copy the examples that log accessToken/API tokens or pass session-transfer tokens in URLs; redact or avoid those patterns. Prefer pinned package/tool versions or a reviewed lockfile, and avoid the printed curl | sh installer unless you independently verify the source and contents.

Vulnerability Patterns
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Findings (5)

T03 · Remote Payload Retrieval and Execution

Error
Location
scripts/utils/validation.mjs:21
Finding

Mutable Remote Installer Is Piped Directly to a Shell

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
references/integration.md:56
Finding

Access Token Is Written to Application Logs

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
references/integration.md:318
Finding

Multi-Resource API Token Is Written to Application Logs

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
references/integration.md:369
Finding

Session-Transfer Credential Is Embedded in a URL Query String

Content
View full analysis
{ const ssoCredentials = await getSSOCredentials(); const webAppUrl = `https://your-web-app.com/login?session_transfer_token=${ssoCredentials.sessionTransferToken}`; await Linking.openURL(webAppUrl); }; ``` ### Technical Analysis The example embeds a session-transfer credential in a URL query parameter. Query strings may be retained or exposed through browser history, application and reverse-proxy access logs, analytics systems, crash reports, screenshots, URL inspection tools, and web application logging. The example does not state that the token must be short-lived and single-use, that the destination must be strictly trusted, or that all participating systems must redact the query parameter. It also does not validate the destination against a fixed HTTPS allowlist. ### Attack Path 1. The application obtains an SSO session-transfer token. 2. It inserts the token into the query string of a web URL. 3. The URL is passed through the operating system and opened by a browser or web application. 4. One or more URL-processing components record the complete URL. 5. An attacker or unauthorized operator obtains the recorded URL. 6. If the credential remains valid and reusable, it is submitted to the intended login endpoint to acquire or transfer the user's session. ### Impact Assessment Successful replay may allow an attacker to assume or transfer the affected user's authenticated session to the associated web application. Exploitability and duration depend on SDK and server controls, particularly whether the token is single-use, narrowly bound, and short-lived. The pr ...[truncated 60 chars]
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
scripts/package.json:9
Finding

Bootstrap Installation Uses Mutable Dependency Ranges Without a Lockfile

Content
View full analysis
/auth0-expo/scripts npm install node bootstrap.mjs ``` The supplied project structure contains no dependency lockfile. ### Technical Analysis Caret version ranges allow npm to select newer compatible releases than those available when the Skill was reviewed. Without a committed lockfile, the exact package graph and integrity hashes are not fixed. Consequently, two users can install different transitive code while following identical Skill instructions. The installation also occurs before the bootstrap accesses the local Expo project and authenticated Auth0 CLI environment. npm package lifecycle scripts are normally permitted during installation, increasing the consequences of a compromised direct or transitive package. No evidence shows that the named packages are currently malicious. The vulnerability is the mutable, unaudited dependency resolution and execution process. ### Attack Path 1. A malicious release is published within an allowed version range, or a direct/transitive dependency is compromised. 2. The user follows the documented `npm install` instruction. 3. npm resolves the compromised version because no lockfile fixes the reviewed dependency graph. 4. Malicious lifecycle or runtime code executes under the user's account. 5. The code accesses resources available to the install or bootstrap process, potentially including project files and local authentication material. ### Impact Assessment A compromised package could execute arbitra ...[truncated 410 chars]
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Rogue AgentSelf-Modification, Session Persistence
Findings (54)

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · references/api.md (reported line 33)May include surrounding context.

md
| Parameter | Type | Description |
|-----------|------|-------------|
| `scope` | `string` | OAuth scopes (default: `openid profile email`) |
| `audience` | `string` | API identifier for access token |
| `organization` | `string` | Organization ID for enterprise login |
| `invitationUrl` | `string` | Organization invitation URL |
| `connection` | `string` | Force a specific connection (e.g., `google-oauth2`) |

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · references/api.md (reported line 57)May include surrounding context.

md
| Parameter | Type | Description |
|-----------|------|-------------|
| `scope` | `string` | OAuth scopes (default: `openid profile email`) |
| `audience` | `string` | API identifier for access token |
| `organization` | `string` | Organization ID for enterprise login |
| `invitationUrl` | `string` | Organization invitation URL |
| `connection` | `string` | Force a specific connection (e.g., `google-oauth2`) |

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · references/integration.md (reported line 72)May include surrounding context.

md
| Parameter | Type | Description |
|-----------|------|-------------|
| `scope` | `string` | OAuth scopes (default: `openid profile email`) |
| `audience` | `string` | API identifier for access token |
| `organization` | `string` | Organization ID for enterprise login |
| `invitationUrl` | `string` | Organization invitation URL |
| `connection` | `string` | Force a specific connection (e.g., `google-oauth2`) |

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · references/api.md (reported line 98)May include surrounding context.

md
| Property | Type | Description |
|----------|------|-------------|
| `accessToken` | `string` | Access token for API calls |
| `idToken` | `string` | ID token with user claims |
| `refreshToken` | `string` | Refresh token (if `offline_access` requested) |
| `tokenType` | `string` | Token type (`Bearer` or `DPoP`) |

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · references/api.md (reported line 174)May include surrounding context.

md
### Token Handling Best Practices

- Never log tokens to the console in production builds
- Use `getCredentials()` to access tokens — it auto-refreshes expired tokens
- Request `offline_access` scope for refresh token support
- Do not store tokens in React state — use `getCredentials()` on demand
- Enable DPoP for enhanced token security (enabled by default)

Credential Access

High
Category
Privilege Escalation
Confidence
97% confidence
Finding

The finding is valid because the example accesses and immediately prints a live access token. In an Expo/mobile auth integration context, developers may copy-paste this pattern into production, turning a sensitive credential into log data that can be replayed to access protected APIs.

Content

Scanner excerpt · references/integration.md (reported line 67)May include surrounding context.

{ scope: 'openid profile email' }, { customScheme: 'auth0sample' } ); console.log('Access Token:', credentials.accessToken);

text

### Login with Audience (API Access)

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · references/integration.md (reported line 168)May include surrounding context.

Credential Auto-Refresh

The credentials manager automatically refreshes expired access tokens using the refresh token. Ensure you request the offline_access scope during login:

typescript
await authorize(

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · references/integration.md (reported line 213)May include surrounding context.

md
domain="YOUR_AUTH0_DOMAIN"
      clientId="YOUR_AUTH0_CLIENT_ID"
      localAuthenticationOptions={{
        title: 'Authenticate to access credentials',
        subtitle: 'Please verify your identity',
        cancelTitle: 'Cancel',
        evaluationPolicy: LocalAuthenticationStrategy.deviceOwnerWithBiometrics,

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · references/integration.md (reported line 252)May include surrounding context.

md
domain="YOUR_AUTH0_DOMAIN"
      clientId="YOUR_AUTH0_CLIENT_ID"
      localAuthenticationOptions={{
        title: 'Authenticate to access credentials',
        subtitle: 'Please verify your identity',
        cancelTitle: 'Cancel',
        evaluationPolicy: LocalAuthenticationStrategy.deviceOwnerWithBiometrics,

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · references/integration.md (reported line 318)May include surrounding context.

Multi-Resource Refresh Tokens (MRRT)

Access tokens for multiple APIs using a single refresh token:

typescript
const { authorize, getApiCredentials, clearApiCredentials } = useAuth0();

Missing User Warnings

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The SSO example appends a sessionTransferToken to a URL query string, which can leak through browser history, referrer headers, analytics, server logs, proxy logs, screenshots, or deep-link handlers. A leaked session transfer token may let an attacker bootstrap an authenticated web session for the victim.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
86% confidence
Finding

The skill instructs use of npx expo install expo-dev-client, which resolves and executes the latest matching package tooling at runtime rather than a pinned version. In an agentic or automated environment, this weakens supply-chain integrity and reproducibility, and could expose users to unexpected or compromised upstream releases.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
86% confidence
Finding

npx expo install react-native-auth0 relies on whatever Expo CLI version npx fetches at execution time. That creates a supply-chain and integrity risk because automated agents may download and run new code without review, and behavior may differ across runs.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
88% confidence
Finding

npx expo prebuild --clean executes Expo CLI code that may be fetched dynamically if not already installed locally. In a skill that encourages automated build/test execution, unpinned tooling increases risk of supply-chain compromise and non-reproducible native project generation.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
88% confidence
Finding

npx expo run:ios can cause the agent to fetch and run an unpinned Expo CLI version at build time. Because this skill explicitly directs the agent to build and iterate on failures, the attack surface is larger than a passive doc example: upstream changes or compromise could directly affect the execution environment.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
88% confidence
Finding

npx expo run:android has the same dynamic-resolution problem as other npx expo commands in the file. In CI or autonomous agent workflows, executing unpinned build tooling can result in unreviewed code execution and inconsistent outcomes.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
79% confidence
Finding

A third matched npx expo token appears on the same troubleshooting line. Repeated examples amplify the chance that automated systems or users execute an unpinned command, increasing practical exposure even if intent is documentation only.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
87% confidence
Finding

The guidance to rerun npx expo prebuild --clean repeats execution of unpinned tooling during remediation steps. Because this may occur multiple times in iterative build-fix loops, the skill compounds the dependency on dynamically fetched code.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
75% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · references/api.md (reported line 78)May include surrounding context.

md
| `customScheme` | `string` | No | Custom URL scheme (lowercase, no special chars). If `"https"`, enables Android App Links with `autoVerify`. |

**What the plugin configures automatically:**
- **iOS**: Adds URL scheme to Info.plist (`CFBundleURLSchemes`), adds deep linking handler to AppDelegate
- **Android**: Sets `manifestPlaceholders` (`auth0Domain`, `auth0Scheme`) in build.gradle

## User Profile Claims

Session Persistence

Medium
Category
Rogue Agent
Confidence
75% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · references/setup.md (reported line 155)May include surrounding context.

md
| `customScheme` | `string` | No | Custom URL scheme (lowercase, no special chars). If `"https"`, enables Android App Links with `autoVerify`. |

**What the plugin configures automatically:**
- **iOS**: Adds URL scheme to Info.plist (`CFBundleURLSchemes`), adds deep linking handler to AppDelegate
- **Android**: Sets `manifestPlaceholders` (`auth0Domain`, `auth0Scheme`) in build.gradle

## User Profile Claims

Session Persistence

Medium
Category
Rogue Agent
Confidence
75% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · references/setup.md (reported line 296)May include surrounding context.

md
| `customScheme` | `string` | No | Custom URL scheme (lowercase, no special chars). If `"https"`, enables Android App Links with `autoVerify`. |

**What the plugin configures automatically:**
- **iOS**: Adds URL scheme to Info.plist (`CFBundleURLSchemes`), adds deep linking handler to AppDelegate
- **Android**: Sets `manifestPlaceholders` (`auth0Domain`, `auth0Scheme`) in build.gradle

## User Profile Claims

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The documentation explicitly logs credentials.accessToken to the console, which normalizes exposing bearer tokens in app logs. On mobile platforms, logs may be accessible during debugging, through crash reporting, or on rooted/jailbroken devices, allowing token replay against APIs.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · references/integration.md (reported line 234)May include surrounding context.

md
| Policy | Behavior |
|--------|----------|
| `BiometricPolicy.default` | System-managed. May skip prompt if recently authenticated. |
| `BiometricPolicy.always` | Always prompts for biometric on every credential access. |
| `BiometricPolicy.session` | Prompts once per session; reuses for the specified timeout. |
| `BiometricPolicy.appLifecycle` | Prompts once until app restarts or credentials are cleared. |

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/integration.md (reported line 286)May include surrounding context.

md
const callApi = async () => {
  const credentials = await getCredentials();
  const headers = await getDPoPHeaders({
    url: 'https://api.example.com/data',
    method: 'GET',
    accessToken: credentials.accessToken,
    tokenType: credentials.tokenType,

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/integration.md (reported line 292)May include surrounding context.

md
const callApi = async () => {
  const credentials = await getCredentials();
  const headers = await getDPoPHeaders({
    url: 'https://api.example.com/data',
    method: 'GET',
    accessToken: credentials.accessToken,
    tokenType: credentials.tokenType,