linkfox-shopee-store-top-picks
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from the Shopee Open API (such as product titles or collection names) and outputs it directly to the agent's context via the
emit_resultfunction inscripts/_shopee_top_picks_common.py. This creates a vulnerability surface for indirect prompt injection if the external data contains instructions designed to manipulate the agent. - Ingestion points: Data is fetched from Shopee's API via the
tool-gateway.linkfox.comproxy and printed to stdout. - Boundary markers: The skill lacks boundary markers or instructions to the agent to ignore potentially malicious content within the API responses.
- Capability inventory: The skill has file system write access (
scripts/_shopee_top_picks_common.py), network communication capabilities (scripts/_shopee_top_picks_common.py,scripts/onboarding.py), and the ability to execute subprocesses (scripts/_shopee_top_picks_common.py). - Sanitization: No sanitization or filtering is performed on the API data before it enters the LLM's context.
- [COMMAND_EXECUTION]: The helper script
scripts/_shopee_top_picks_common.pyusessubprocess.runto execute a local dependency checker script (scripts/check_auth_dependency.py). While used legitimately for verifying the installation of necessary components, it constitutes local command execution that could be targeted if the local paths were compromised.
Audit Metadata