The Agent Skills Directory

[SAFE]: The skill consists entirely of natural language instructions for writing marketing copy and contains no executable scripts, commands, or binary files.
[SAFE]: No obfuscation, hardcoded credentials, or data exfiltration patterns were identified.
[PROMPT_INJECTION]: The skill exhibits an Indirect Prompt Injection surface (Category 8) as it processes external context. Ingestion points: User context provided via the default prompt in agents/openai.yaml. Boundary markers: None identified. Capability inventory: No file system writes, network operations, or subprocess calls identified in the instructions. Sanitization: The skill instructs the agent to never invent claims and to keep external writes behind explicit human approval.

write-copy