build-tcg-deck

Pass

Audited by Gen Agent Trust Hub on Mar 18, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: No evidence of prompt injection or behavior override instructions. The language used is purely instructional for the intended TCG deck-building task.
  • [DATA_EXPOSURE]: No hardcoded credentials, API keys, or access to sensitive file paths were detected. The skill does not attempt to access private system data.
  • [REMOTE_CODE_EXECUTION]: The skill does not include any commands to download, execute, or interpret remote code or scripts.
  • [OBFUSCATION]: No obfuscated text, encoded strings, or hidden content (such as zero-width characters) were found in the file.
  • [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it ingests external meta-game data via WebSearch and WebFetch tools. However, the risk is negligible because the skill lacks the capability to perform dangerous actions (like file modification or command execution) based on the ingested data.
  • Ingestion points: WebSearch and WebFetch are used in Step 4 to retrieve tournament results and tier lists from external websites.
  • Boundary markers: None identified.
  • Capability inventory: The skill is limited to natural language processing and deck list generation; it has no access to subprocesses, eval(), or file-system write operations.
  • Sanitization: None identified, but the structured procedure naturally constrains the model's focus to card-game data.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 18, 2026, 07:12 AM