The Agent Skills Directory

[SAFE]: No security issues detected. The skill instructions are focused on trademark analysis and use appropriate tools for their stated purpose.- [EXTERNAL_DOWNLOADS]: The skill references official trademark registries (WIPO, USPTO, EUIPO). These are well-known services and their use for data retrieval is safe and consistent with the skill's functionality.- [DATA_EXFILTRATION]: No hardcoded credentials or access to sensitive system files were found.- [PROMPT_INJECTION]: The instructions do not contain any patterns attempting to bypass agent safety guidelines or override system prompts. Additionally, the skill processes data from external registries (ingestion points: WebFetch/WebSearch in Step 2 and 5; capabilities: WebFetch, WebSearch, Read, Grep, Glob; boundary markers and sanitization: absent). While this is a theoretical indirect prompt injection surface, the risk is negligible given the trusted nature of the sources.

screen-trademark