The Agent Skills Directory

[SAFE]: The skill's logic is entirely instructional and conversational. It does not request any tool permissions, execute shell commands, or perform network operations. All image processing described refers to the agent's internal capabilities or external image generation tools using the provided prompts.
[PROMPT_INJECTION]: The skill is designed to ingest and summarize user-provided articles, scripts, and topics to generate cover titles. This creates a standard surface for indirect prompt injection where adversarial text in the user's content could attempt to influence the agent. However, this is part of the skill's primary functionality and is categorized as safe within that context.
Ingestion points: Reads user-provided content (articles, scripts, or topics) as specified in the SKILL.md workflow.
Boundary markers: The instructions do not specify any delimiters (e.g., XML tags or triple quotes) to separate user input from system instructions.
Capability inventory: The skill's capabilities are limited to conversational interaction and the generation of text-based prompts.
Sanitization: No explicit sanitization or filtering of the user-provided content is implemented within the skill instructions.

atutun-xhs-cover