The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute a suite of local Python scripts (gen_image.py, caption_composite.py, safe_zones.py, schedule_post.py) for core operational tasks including image generation, image editing, text compositing, and batch scheduling via the Postiz API.
[INDIRECT_PROMPT_INJECTION]: The skill acts as a downstream consumer of data produced by a separate trendwatch skill, creating a surface for indirect instructions to influence the generation and scheduling process.
Ingestion points: Reads hook ideas and content formats from external files like ideas/YYYY-MM-DD.md and current-plan.md created by other skills.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when processing data from the trendwatch skill for inclusion in captions or prompts.
Capability inventory: The skill has access to Bash for script execution, Write for file creation, and Agent for inter-agent communication, which could be leveraged if malicious instructions were successfully injected into the source data.
Sanitization: There is no mention of sanitizing or validating the content retrieved from external markdown files before it is interpolated into image generation prompts or PIL-based caption overlays.

carousel-conveyor