casualize-names

Pass

Audited by Gen Agent Trust Hub on Mar 27, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to Indirect Prompt Injection (Category 8) because it interpolates untrusted data from an external source directly into AI prompts.
  • Ingestion points: Untrusted data enters the agent context via worksheet.get_all_values() in all scripts (e.g., scripts/casualize_batch.py).
  • Boundary markers: The skill does not use any boundary markers or delimiters to isolate the spreadsheet content from the system instructions.
  • Capability inventory: The skill possesses the ability to perform network operations (Anthropic API) and modify external files/data (Google Sheets via worksheet.batch_update).
  • Sanitization: No escaping, validation, or filtering of the external content from the spreadsheet is performed before interpolation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 27, 2026, 12:42 AM
Security Audit — agent-trust-hub — casualize-names