appmigrationkit

Warn

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: MEDIUMMETADATA_POISONINGINDIRECT_PROMPT_INJECTION
Full Analysis
  • [METADATA_POISONING]: The skill documentation refers to fictitious software versions (iOS 26.0+) and utilizes an unofficial domain (sosumi.ai) for all technical documentation links instead of official developer resources. This misleading information could cause developers or agents to rely on non-existent APIs or untrusted documentation sources.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides implementation patterns for importing external data from untrusted sources into the application's secure container and shared app groups. This functionality lacks robust content validation, creating a potential vector for malicious data to influence the system.
  • Ingestion points: The importResources method in references/appmigrationkit-patterns.md and references/core-implementation.md accepts an external URL to imported resources.
  • Boundary markers: No delimiters or isolation techniques are recommended for the ingested data to prevent it from being interpreted as instructions.
  • Capability inventory: The skill utilizes FileManager to perform write operations within the app container and security application groups.
  • Sanitization: The provided code snippets perform direct file copying operations without verifying the integrity or safety of the ingested content.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 11, 2026, 11:13 AM
Security Audit — agent-trust-hub — appmigrationkit