The Agent Skills Directory

[SAFE]: Analysis of the skill instructions, metadata, and logic found no evidence of malicious behavior, obfuscation, or unauthorized access. The skill follows best practices for generating marketing copy. It utilizes the WebSearch tool to ingest untrusted data from the internet to customize ad content. This ingestion surface was evaluated for indirect prompt injection risk and found to be safe: search results enter the context without explicit boundary markers or sanitization, but the agent's capabilities are limited to generating a local markdown file, preventing potential downstream exploitation.

restaurant-ads