The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions to install the belt CLI tool and related skills from the vendor's repository and the belt-sh NPM package. These resources are part of the vendor's supported ecosystem.
[COMMAND_EXECUTION]: The skill uses the belt CLI to perform audio generation and manipulation. The use of the shell is restricted to the belt command via the skill's frontmatter configuration, which limits the potential for unauthorized command execution.
[PROMPT_INJECTION]: The skill features workflows that ingest external content, such as documents for "NotebookLM-style" podcast generation, into LLM prompts. This is a common pattern for content generation skills.
Ingestion points: User-provided document content and generated script segments are interpolated into prompts for the Claude LLM.
Boundary markers: No specific delimiters or boundary instructions are used in the prompt templates to separate user data from instructions.
Capability inventory: The skill possesses the capability to execute shell commands via the belt CLI.
Sanitization: No explicit sanitization or filtering of the user-provided data is documented before it is passed to the LLM.

ai-podcast-creation