Static analysis reported command, network, storage, filesystem, environment, and prompt-injection indicators, but review found documentation examples for Expo and React Native development. No prompt injection text, credential exfiltration, automatic command execution, malicious network request, or hidden payload behavior was found.
Static analysis reported command, network, storage, filesystem, environment, and prompt-injection indicators, but review found documentation examples for Expo and React Native development. No prompt injection text, credential exfiltration, automatic command execution, malicious network request, or hidden payload behavior was found.
Static analysis reported many command, network, storage, filesystem, and prompt-injection indicators, but review found they are Markdown documentation examples for Expo and React Native development. No prompt injection text, credential exfiltration, executable installer, or malicious automation was found; publication is acceptable with low risk because the skill teaches app code that may request permissions or use local storage.
Storage Examples Are App Guidance, Not Credential Access
The storage reference teaches localStorage and SQLite examples and explicitly directs sensitive data such as tokens and passwords to expo-secure-store. This is not evidence of marketplace credential theft.
The examples are framed as app data storage patterns, and the sensitive-data row recommends secure storage. The risk is limited to developers applying the examples incorrectly.
Capability review items (2)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
The dynamic import finding is a React.lazy example for loading a local scene component. It does not import from user-controlled input or a remote URL.
The import target is a static local alias and the surrounding section describes code splitting. There is no dynamic expression built from untrusted data.
Environment And Path Findings Are Benign Framework Examples
The environment access is a recommendation to use Expo platform metadata, and the path traversal alert is a normal relative import. Neither indicates secret access or filesystem traversal.
The environment variable name is EXPO_OS, not a secret, and the relative import targets an application component. No file reads, writes, or traversal over user input were found.
Static false positives ignored (3)
These static matches were dismissed by semantic review or matched schema-only tokens, so they are shown for transparency but do not drive the quality score.
Static Command Execution Findings Are Markdown False Positives
The reported Ruby or shell backtick execution locations are Markdown fences, inline code, and Expo command examples. They are not executable skill code and do not create command injection by themselves.
The sampled locations show Markdown and TypeScript documentation context, not Ruby backtick execution. The repeated detections align with code fences and inline examples across the reference files.
The network indicators point to an Apple documentation link and example audio or video sources. I found no evidence that the skill sends user data, credentials, or telemetry to those URLs.
The URLs appear in user-facing documentation or sample playback code. No surrounding code indicates exfiltration, credential handling, or background network activity.
No Prompt Injection Attempt Found In Reviewed Matches
The scanner flagged a system-prompt pattern, but the reviewed line contains PlatformColor("systemBackground") inside a React Native view style. No override, skip-audit, or system-message instruction was found.
The suspicious token is part of an iOS color constant in UI code. A broader phrase search found no explicit prompt injection language in SKILL.md or the reference files.
All 614 static findings are false positives. This is legitimate Expo documentation containing code examples for building native iOS apps. The scanner incorrectly flagged documentation patterns including CLI command examples (npx expo), CSS color values (rgba), and regex patterns as security issues. No malicious code present.
13
Files scanned
3,329
Lines analyzed
3
Review items
7
False positives ignored
Static false positives ignored (7)
These static matches were dismissed by semantic review or matched schema-only tokens, so they are shown for transparency but do not drive the quality score.
Documentation Code Examples Misidentified as Shell Commands
The static scanner flagged 547 instances of 'Ruby/shell backtick execution' in markdown files. These are legitimate code examples showing CLI commands like `npx expo start` and `npx expo run:ios` in documentation. The backticks are markdown code formatting, not shell execution.
Backticks in markdown denote inline code formatting. The content describes Expo CLI commands for development workflows, not shell execution.
CSS Color Values Misidentified as Cryptographic Patterns
The scanner flagged 'weak cryptographic algorithm' on CSS color values like `rgba(0, 0, 0, 0.05)` and regex patterns like `match(/\((.*)\)/)`. These are standard CSS and JavaScript patterns for styling and string parsing, not cryptographic operations.
RGBA color values are CSS syntax for transparent colors. The regex extracts route segments for navigation logic, unrelated to cryptography.
Storage API Documentation Misidentified as Security Risk
The scanner flagged references to AsyncStorage, SecureStore, and SQLite as 'sensitive data access'. These are legitimate Expo storage APIs being documented for app data persistence patterns.
The skill documents Expo's official storage APIs for persistence, a standard requirement for mobile app development.
The scanner flagged line 32 as 'template injection targeting system prompt'. This line reads: 'Start with Expo Go: Run `npx expo start` and scan the QR code with Expo Go' - pure documentation text.
This is plain text documentation describing the Expo Go development workflow. No template injection mechanism exists.
Heuristic Alert Triggered by Documentation Patterns
Critical heuristic alerts flagged 'dangerous combination: code execution + network + credential access'. These patterns co-occur because the documentation covers complete app development including networking (expo-camera, expo-media-library) and storage (SecureStore) - standard mobile app patterns.
Mobile app development legitimately requires networking, storage, and command-line tooling. Documentation covering these topics triggers pattern-based heuristics incorrectly.
Documentation skill containing markdown guides with code examples. Static scanner flagged pattern-based false positives triggered by markdown code blocks (backticks), legitimate documentation URLs, environment variable references, and file path examples. All findings are false positives - the skill contains no executable code, only documentation for Expo Router development.
13
Files scanned
3,329
Lines analyzed
5
Review items
4
False positives ignored
Static false positives ignored (4)
These static matches were dismissed by semantic review or matched schema-only tokens, so they are shown for transparency but do not drive the quality score.
Static scanner detected backtick syntax in markdown code blocks. This is normal documentation formatting for code examples, not shell command execution. All instances are within fenced code blocks (```tsx, ```) used for React Native/Expo code examples.
All backtick occurrences are within markdown fenced code blocks showing example Expo/React Native code. No shell command execution exists in this documentation skill.