raffle-winner-picker
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from untrusted external sources, which could contain embedded malicious instructions.
- Ingestion points: Data is read from user-provided local files (e.g.,
entries.csv,contest-entries.xlsx) and external URLs (Google Sheets). - Boundary markers: The skill instructions do not provide delimiters or "ignore instructions" warnings to prevent the agent from obeying commands hidden within the spreadsheet data.
- Capability inventory: The skill performs file reading, data processing, and potentially data exporting based on the contents of external files.
- Sanitization: No sanitization or validation of the input data is defined to filter out potential prompt injection attempts within the entries list.
Audit Metadata