The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized data access were detected. The skill's instructions and references align with its stated purpose of providing design guidelines and component templates.
[PROMPT_INJECTION]: While the skill ingests external data from JSON files (copy.json and design-brief.json) for code generation, this is an architectural necessity of its function as a design maestro. The processing is restricted to UI component interpolation, and no attempts to bypass safety filters or override agent behavior were found.
[DATA_EXFILTRATION]: The skill uses the Read tool to access local context files provided by previous agents in the workflow. This behavior is documented and legitimate for inter-agent communication within the execution environment. No evidence of network exfiltration or sensitive credential harvesting was observed.

yahav-design-agent