The Agent Skills Directory

[SAFE]: The skill contains purely instructional content and UI design patterns for ecommerce scenarios. No malicious patterns, command executions, or network exfiltration attempts were detected.
[INDIRECT_PROMPT_INJECTION]: The skill instructions specify that the agent should read local strategy files (e.g., .agents/marketing-strategy.md) to provide contextual advice.
Ingestion points: Reads local files like marketing-strategy.md, marketing-learnings.md, and brand-tier-guide.md if present in the workspace.
Boundary markers: None specified in the instructions.
Capability inventory: No dangerous capabilities (file writing, network operations, or subprocess execution) are present in the skill's logic.
Sanitization: None specified for the ingested content. However, since the skill only outputs marketing and design recommendations, the risk is minimal.

paywalls