The Agent Skills Directory

[PROMPT_INJECTION]: No patterns of prompt injection, role-play overrides, or instructions to bypass safety guidelines were found. The skill instructions remain strictly within the bounds of frontend development tasks.- [DATA_EXPOSURE_AND_EXFILTRATION]: No sensitive data access, hardcoded credentials, or unauthorized network communication patterns were identified. There are no references to sensitive environment variables or local credential stores.- [REMOTE_CODE_EXECUTION]: The skill does not contain instructions to download or execute code from external sources. No package installation commands or remote script piping (e.g., curl|bash) were detected.- [OBFUSCATION]: No evidence of Base64 encoding, zero-width characters, homoglyphs, or other obfuscation techniques was found in the analyzed files. All text and code are presented in clear, human-readable formats.- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided inputs such as bug reports and user journeys to inform UI changes. While this represents a data ingestion surface, the instructions do not demonstrate unsafe interpolation or a lack of boundaries that would typically facilitate indirect prompt injection attacks.

frontend-dev