The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run local commands via node bin/mdpr-skill.js to perform specialized presentation review tasks. These operations are restricted to the local workspace and are consistent with the skill's stated purpose of enhancing the MDPR tool workflow.
[PROMPT_INJECTION]: The skill is designed to ingest and analyze user-controlled files like Markdown decks and manifest JSONs, creating a surface for indirect prompt injection.
Ingestion points: Files like deck.md, dist/mdpresent-manifest.json, and template-layout-catalog.json are read from the local workspace.
Boundary markers: There are no explicit instructions for the agent to use boundary markers or delimiters when processing these external files to prevent instruction leakage.
Capability inventory: The skill enables the agent to execute local shell commands through Node.js scripts based on the analysis of these files.
Sanitization: No sanitization or validation routines for the external file content are described within the skill instructions.

mdpr-skill