The Agent Skills Directory

[PROMPT_INJECTION]: No instructions were found that attempt to override agent behavior or bypass safety guidelines.
[DATA_EXPOSURE_AND_EXFILTRATION]: No sensitive file access, credential exposure, or unauthorized network requests were identified.
[OBFUSCATION]: No Base64, zero-width characters, or hidden URLs were detected in the instructions or examples.
[UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill does not perform any external downloads or execute remote scripts. All code examples are static and instructional.
[PRIVILEGE_ESCALATION]: No commands for acquiring elevated system permissions or bypassing access controls were found.
[PERSISTENCE_MECHANISMS]: No methods for maintaining unauthorized access across sessions or modifying system startup files were detected.
[METADATA_POISONING]: The skill's metadata is descriptive and consistent with its UI layout auditing functionality.
[INDIRECT_PROMPT_INJECTION]: The skill analyzes layout in frontend code but does not include instructions that would lead to dangerous tool invocation or data exfiltration.
[TIME_DELAYED_ATTACKS]: No logic gating behavior behind time, date, or environment-specific conditions was identified.
[DYNAMIC_EXECUTION]: No runtime code generation, injection, or unsafe data deserialization patterns are present.
[DYNAMIC_CONTEXT_INJECTION]: No instances of shell command execution at load time using the ! syntax were identified.

better-layout