The Agent Skills Directory

[SAFE]: No malicious patterns or security issues were detected. The skill's instructions and references are consistent with its stated purpose as a game design planning tool.
[DATA_EXPOSURE_EXFILTRATION]: The skill uses web research capabilities to gather market data and design references from public sources (e.g., Steam, Metacritic, GDC Vault). There is no indication of access to sensitive local files or environment variables.
[PROMPT_INJECTION]: The instructions follow standard role-based constraints and refer to external boundary files correctly. No attempts to bypass safety filters or override system instructions were found.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from the web (community forums, store pages). While this represents an ingestion surface, the skill is restricted to producing markdown documentation and balance math, and it is explicitly prohibited from writing or executing implementation code, which significantly mitigates potential injection risks.
[REMOTE_CODE_EXECUTION]: The skill explicitly states 'Does not write code' and 'Never write implementation code,' effectively eliminating common execution vectors.

quest