The Agent Skills Directory

[Prompt Injection] (SAFE): No instructions attempting to bypass safety filters, override agent behavior, or extract system prompts were detected in the documentation or metadata.- [Data Exposure & Exfiltration] (SAFE): The skill does not perform any network operations, access sensitive local files, or contain hardcoded credentials.- [Unverifiable Dependencies & Remote Code Execution] (SAFE): No scripts or package management files (like requirements.txt or package.json) are included. The README explicitly states there are no external dependencies.- [Indirect Prompt Injection] (LOW): The skill possesses a surface for indirect prompt injection as it processes user-provided context for ideation (e.g., team size, problem descriptions). However, because the skill has no executable capabilities, network access, or file-writing permissions, the risk of an exploit is negligible.- [Persistence & Privilege Escalation] (SAFE): No commands or configurations related to system persistence or unauthorized privilege elevation were found.

ideaspark-navigator