The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and transform untrusted prose provided by the user into D&D-style narration.
Ingestion points: The skill reads user-supplied text as its primary input, as defined in the workflow of SKILL.md and demonstrated in references/examples.md.
Boundary markers: The instructions in SKILL.md (Workflow Step 4) direct the agent to partition mixed content into Read Aloud, DM Lore, and Mechanics sections. While this helps organize output, it does not programmatically isolate the input text from the agent's instructions.
Capability inventory: The skill does not have access to sensitive tools; it performs text transformation only and does not include network, file-system write, or shell execution capabilities.
Sanitization: The "Finish gate" in SKILL.md provides a logical safety check, requiring the agent to verify that the transformed prose preserves the facts and mechanics of the original source, which mitigates the impact of instructions embedded within the user data.
[EXTERNAL_DOWNLOADS]: The skill's documentation contains references to external resources and installation instructions.
Reference links: references/sources.md includes links to the official System Reference Document (SRD) on dndbeyond.com and media.dndbeyond.com. These are recognized as official and well-known services for the skill's context.
Installation method: The README.md file provides a standard installation command using npx for the author's skill repository (SirDarcanos/agentskills).

dnd-format-prose