The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because its core function is to process potentially untrusted external data.
Ingestion points: In SKILL.md, the instructions require the agent to work from sources like "blog post, transcript, docs" to generate the script.
Boundary markers: The skill does not provide specific delimiters or instructions to the agent to distinguish between the user's source material and the agent's own instructions, nor does it warn the agent to ignore instructions embedded in the source text.
Capability inventory: The skill's capabilities are limited to text generation (scripts, titles, and descriptions). There are no file system modifications, command executions, or network operations requested within the instruction set.
Sanitization: There are no guidelines for sanitizing or escaping the external content before it is incorporated into the script-writing process.
[EXTERNAL_DOWNLOADS]: The skill references an external repository owned by the author.
Evidence: The file SKILL.md contains a link to https://github.com/mohitagw15856/ContentGoldMine.
Context: This link points to a resource owned by the skill's author, intended for automating the script generation workflow described in the skill.

youtube-script