gzh-astock-top
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill performs network requests to the vendor's API at https://redfox.hk to fetch financial news and account data. This domain is managed by the skill's author, redfox-data. The requests include the user's API key for authentication and specific account names for data retrieval. This behavior is standard for the skill's primary functionality.
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves article titles and URLs from an external API and instructs the agent to generate summaries based on this content. This creates a surface where malicious instructions embedded in article headlines could potentially influence the agent's behavior.
- Ingestion points: The scripts scripts/fetch_astock_accounts.py and scripts/fetch_subscribed_updates.py retrieve article titles from the redfox.hk API.
- Boundary markers: The instructions in SKILL.md do not include specific delimiters or warnings to ignore instructions embedded in the external content.
- Capability inventory: The skill has capabilities to perform network requests and write to local configuration and cache files (subscriptions.json and the cache/ directory).
- Sanitization: There is no evidence of sanitization or filtering of the article titles before they are processed by the agent for summary generation.
Audit Metadata