This report does not block or authorize the manifest or ZIP.
Static analysis flagged many Markdown code fences as shell execution and misread ordinary Expo examples as crypto, reconnaissance, or credential access. Manual review found no prompt injection, obfuscation, hidden execution, or malicious exfiltration intent; the only real network example is a documented Expo Push API call.
Report position
Historical report
Open audit history before using this report to install.
Audit attestation
Not attestable
The required immutable binding is incomplete.
Human verification
Not verified
No human verification is recorded for this report.
Coverage
1 Files scanned · 466 Lines analyzed
1 item shown for review
Limitations
This report does not claim runtime or sandbox execution and does not prove the absence of side effects.
01
Evidence chain
Follow the evidence from source binding to the install contract. Available evidence supports verification; it is not a safety guarantee.
The hardcoded URL is the Expo Push API endpoint in a backend notification example. It sends caller-provided notification fields to Expo, not hidden marketplace user data.
The network call is real, but the surrounding section clearly labels it as a push notification example. No covert destination or credential harvesting behavior is present.
04
Expert evidence
Immutable subject identity, scanner metadata, dismissed matches, and source-level evidence.
Artifact subject
Marketplace commit
Unavailable
Content hash
Unavailable
Tree hash
Unavailable
Skill path
Unavailable
Audit payload hash
Unavailable
Analysis metadata
Audit model: codex
Analysis state: Complete
Scope is limited to the recorded files, lines, methods, and evidence. No runtime or sandbox execution is claimed.
Markdown Code Fences Misclassified as Shell Execution
The external command detections point to Markdown fenced code examples. These snippets are educational TypeScript or Python examples, not executable skill hooks or shell commands.
Each location is a Markdown fence boundary or nearby example code. No instruction in SKILL.md asks the assistant to run these snippets automatically.
Mobile Storage Guidance Is Not Browser Credential Theft
The storage section shows AsyncStorage for preferences and SecureStore for tokens. This is expected React Native guidance and does not access browser storage.
The detected storage content is mobile app storage documentation. SecureStore is recommended for sensitive data instead of unsafe plain storage.
No weak cryptographic algorithm is documented or invoked at the flagged lines. The lines contain skill metadata and keyboard listener cleanup code.
Line 3 is a natural language description, and lines 446 and 449 remove keyboard listeners. No MD5, SHA-1, DES, or similar weak crypto use appears there.
System Reconnaissance Detection Is Keyboard UI Guidance
The flagged line introduces a KeyboardAvoidingView example. It does not enumerate files, environment variables, devices, users, or operating system details.
The context is React Native keyboard layout handling. There is no system information collection or reconnaissance behavior in the surrounding code.
Critical Combination Heuristic Dismissed After Review
The critical heuristic combined unrelated false positives. Manual review found no code execution capability, no credential exfiltration, and no obfuscated control flow.
The file is a single Markdown guide with visible examples. The supposed execution, credential, and crypto signals do not form a real attack chain.