The Agent Skills Directory

[PROMPT_INJECTION]: The skill provides structured instructions for a specific task (AI SEO) and does not attempt to override system prompts, bypass safety filters, or extract underlying instructions.
[DATA_EXPOSURE]: No sensitive file paths, hardcoded credentials, or unauthorized network operations were detected. The reference to .agents/product-marketing-context.md is a standard practice for maintaining task context within an agentic workflow.
[EXTERNAL_DOWNLOADS]: The skill refers to well-known services (Google, Perplexity, OpenAI, Anthropic, Brave) and official documentation for those services. No suspicious or unverified third-party downloads are present.
[REMOTE_CODE_EXECUTION]: There are no executable scripts, shell commands, or dynamic code execution patterns in the provided files. The skill is entirely descriptive and instructional markdown.
[OBFUSCATION]: No Base64, zero-width characters, or other encoding techniques were found that would suggest an attempt to hide malicious intent.
[INDIRECT_PROMPT_INJECTION]: While the skill is designed to process user-provided content for optimization purposes, it does not possess the execution capabilities (file writes, network requests, or tool usage) that would enable an indirect injection attack. The lack of tool execution within this skill reduces the risk to a negligible level.

ai-seo