The Agent Skills Directory

[SAFE]: The skill implements professional marketing frameworks with significant emphasis on output quality and data integrity. No malicious instructions, obfuscation, or unauthorized access patterns were detected.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data from web search results and user-provided URLs in its research modes. This is a standard requirement for marketing analysis and is mitigated by the skill's built-in heuristics.
Ingestion points: SKILL.md (SEO_AUDIT, COMPETITIVE, and BRAND modes) and references/seo-audit.md.
Boundary markers: The skill uses strict 'Gate' checkpoints and structured templates to maintain focus on the intended task.
Capability inventory: Utilizes web search and reads user-provided files or text to generate reports.
Sanitization: Incorporates references/llm-marketing-failure-modes.md to identify and prevent the inclusion of fabricated metrics or hallucinated competitor data in the final output.

marketing