apple-app-store-product
Pass
Audited by Gen Agent Trust Hub on Aug 7, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill connects to
https://ai-factory.frevana.comto retrieve App Store metadata. - [COMMAND_EXECUTION]: The skill executes a bash script (
get_apple_product.sh) which in turn usescurlfor network requests andpython3for JSON processing. - [DATA_EXFILTRATION]: The skill manages a
FREVANA_TOKENfor authentication. The token is passed via environment variables or a CLI flag and is used as a Bearer token in the Authorization header of the API request to the vendor's infrastructure. - [COMMAND_EXECUTION]: The script uses inline Python code execution to safely construct JSON payloads and validate API responses.
- [PROMPT_INJECTION]: The skill processes data from an external API (
ai-factory.frevana.com). This data is subsequently summarized or returned to the user, creating a surface where malicious content in the API response could potentially influence the agent's behavior. 1. Ingestion points: Data enters the agent context via the API response fromai-factory.frevana.comwhich is read and summarized by the agent. 2. Boundary markers: The script validates that the input is JSON but does not apply specific delimiters or instructions to ignore embedded commands. 3. Capability inventory: The skill can execute shell commands, write files to the local disk, and make network requests. 4. Sanitization: The script performs JSON schema validation but does not filter the semantic content of the product metadata.
Audit Metadata