The Agent Skills Directory

[PROMPT_INJECTION]: No attempts to override agent behavior, bypass safety filters, or extract system prompts were found. The skill contains standard operational constraints and instructions for marketing document generation.- [DATA_EXPOSURE]: No hardcoded credentials, sensitive file paths, or unauthorized network operations were detected. The skill requests standard business inputs like contact numbers and brand colors, which are appropriate for its purpose.- [OBFUSCATION]: The text was analyzed for Base64, zero-width characters, Unicode tags, and homoglyphs. No malicious obfuscation or hidden content was found.- [EXTERNAL_DOWNLOADS]: References to external services like Linktree, Beacons, ManyChat, and Meta Ads Manager are provided as tactical recommendations for the user. No remote code execution or automated script downloads are present.- [INDIRECT_PROMPT_INJECTION]: The skill acts on user-provided client data to generate a strategy. While it functions as a data processing surface, the lack of powerful tools such as subprocess, eval, or file-writing tools minimizes the risk of indirect injection exploitation.

platform-instagram