The Agent Skills Directory

[PROMPT_INJECTION]: The instructions focus entirely on providing financial simulations and do not contain any patterns intended to bypass safety guidelines or override agent behavior.
[DATA_EXFILTRATION]: No sensitive file paths, hardcoded credentials, or unauthorized network operations were identified. The skill does not attempt to leak user data.
[REMOTE_CODE_EXECUTION]: There are no patterns involving the download or execution of scripts from remote servers. The instruction to use code for calculations refers to the agent's internal code execution tool.
[COMMAND_EXECUTION]: No dangerous system-level commands, privilege escalation attempts, or persistence mechanisms are present.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data via search queries for fund performance and valuations. While this creates a surface for indirect injection, the skill's narrow focus on financial modeling and its instruction to use code for deterministic calculations mitigate this risk to a standard level for search-enabled agents.

fund-calculator