The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data, creating a potential surface for indirect prompt injection.
Ingestion points: Video transcripts (get_transcript) and clip metadata including titles and hooks (list_clips) are retrieved from external videos submitted by users.
Boundary markers: There are no specific instructions or delimiters provided to the agent to treat transcript content as untrusted data or to ignore embedded instructions.
Capability inventory: The skill possesses network capabilities via tools like submit_video and create_upload for handling media files.
Sanitization: The instructions do not define any sanitization or filtering of the content retrieved from external video sources before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill documentation references external packages for installation.
The SKILL.md file suggests installing focused sub-skills using the command npx skills add OpenClip-App/agent-skills, which targets the author's official repository.

openclip