raffle-winner-picker
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external, untrusted sources including Google Sheets, CSV files, and Excel documents provided by the user.
- Ingestion points: Data is imported via URL references to Google Sheets and local file paths (e.g.,
entries.csv,contest-entries.xlsx) provided in user prompts. - Boundary markers: The instructions lack specific guidance for the agent to use delimiters or ignore embedded instructions within the spreadsheet content, increasing the risk that the AI might interpret data as commands.
- Capability inventory: To function, the skill requires the agent to utilize file-reading and web-browsing capabilities to access the content of the winner lists.
- Sanitization: There is no mention of sanitizing, escaping, or validating the input data before it is processed or displayed in the final output.
Audit Metadata