The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected. The skill's behavior is restricted to generating text-based ideation directions and optionally saving them to a local project file.
[PROMPT_INJECTION]: The skill uses instructional markers to define its scope and prevent accidental activation. These are legitimate configuration boundaries rather than attempts to bypass system safety. It takes user-provided ideas as input but lacks exploitable capabilities like network access or code execution that could be targeted by indirect injection.
[DATA_EXFILTRATION]: There are no network requests, credential access, or operations involving sensitive user data. The file operations are confined to the .byte-os directory within the project context for the purpose of saving brainstorm results.
[COMMAND_EXECUTION]: The skill does not contain shell commands, subprocess calls, or any mechanisms for arbitrary code execution.

byte-brainstorm