The Agent Skills Directory

[PROMPT_INJECTION]: The skill directs the agent to analyze external, third-party content including competitor websites, social profiles, and advertisements. This creates a surface for indirect prompt injection, where malicious instructions could be embedded in those external materials to manipulate the agent's analysis or output.
Ingestion points: External URLs, social media profiles, and marketing materials.
Boundary markers: There are no instructions provided to the agent to treat external content as untrusted or to ignore potential instructions embedded within the analyzed materials.
Capability inventory: The skill focus is on analysis and report generation; no dangerous capabilities like arbitrary file writes or network exfiltration are defined within this specific file.
Sanitization: The skill does not define any sanitization or validation steps for the data retrieved from external sources.
[COMMAND_EXECUTION]: The instructions mandate the execution of an external command or skill named brand-init when initial configuration is missing, which involves executing logic outside of this instruction file to set up the environment.

competitor-branding