The Agent Skills Directory

[SAFE]: Comprehensive analysis across all 11 threat categories identified no malicious patterns. The skill consists entirely of instructional Markdown files for orchestrating content marketing specialists.
[INDIRECT_PROMPT_INJECTION]: The skill creates a potential data ingestion surface by instructing agents to process external data sources such as project-specific brand context files, customer success stories, and website content. This risk is assessed as safe/low because the skill does not define any dangerous capabilities (e.g., shell access, network exfiltration) that could be exploited via malicious input.
Ingestion points: File paths such as '.claude/brand-context.md', './docs/brand-context.md', and external URLs for research.
Boundary markers: The skill specifies that a 'Rules for agents reading this file' section in the context file should override an agent's own defaults.
Capability inventory: The skill is restricted to content creation tasks (Markdown, PDF, video scripts) and does not define tools for command execution or remote data transmission.
Sanitization: No specific sanitization or filtering logic is prescribed for the external content ingested.

content-marketing