The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references a variety of external URLs within the references/cases.md file to provide visual and structural reference data.
Evidence: The file includes multiple links to goodcase.ai and media.goodcase.ai for case summaries and finished media examples, as well as links to the original creators on x.com. These references are integral to the skill's primary function of providing style-grounded prompt engineering.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and interpret external data from case studies which could theoretically contain instructions intended to influence the agent.
Ingestion points: The instructions require the agent to read references/cases.md and use browser or image tools to inspect external media linked in the case descriptions.
Boundary markers: There are no explicit delimiters used to isolate external content from the agent's core instructions.
Capability inventory: The skill is limited to reading local reference files, using browser tools for inspection, and generating text-based image prompts. It does not have capabilities for writing to the file system, executing shell commands, or network exfiltration.
Sanitization: No explicit content sanitization or instruction-filtering logic is present for the ingested case data.

character-style-consistency