The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill is a catalogue stub that downloads its full implementation, including scripts and reference files, from an external GitHub repository (LearnPrompt/humanize-ppt) using the npx skills add command.
[COMMAND_EXECUTION]: The instructions direct the agent to execute shell commands to verify installation paths and to perform the installation of the remote skill package.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (notes, transcripts, documents, and web links) to generate presentation briefs and fix instructions, which creates a surface for indirect prompt injection.
Ingestion points: Notes, transcripts, documents, and web links provided by the user as input for presentation planning.
Boundary markers: No specific delimiters or safety instructions regarding untrusted data were found in the provided files.
Capability inventory: Generates production briefs, style briefs, and slide-fix instructions that direct the behavior of downstream agent tools.
Sanitization: There is no evidence of content validation or sanitization for the external data being processed.

humanize-ppt