The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from external sources.
Ingestion points: The workflow fetches XML data from YouTube RSS feeds and retrieves full video transcripts using the mmk youtube transcript tool (SKILL.md).
Boundary markers: There are no defined delimiters or specific instructions provided to the agent to ignore potentially malicious instructions embedded within the fetched transcripts during the summarization phase.
Capability inventory: The skill possesses the ability to perform network operations via curl and write data to an external service using mmk notion database upsert (SKILL.md).
Sanitization: The instructions do not include steps to sanitize or validate the content of the transcripts before they are processed by the AI or written to the Notion database.
[EXTERNAL_DOWNLOADS]: The skill performs network requests to fetch external data.
Evidence: Uses curl to download RSS feed content from https://www.youtube.com/feeds/videos.xml (SKILL.md).
Context: This operation targets a well-known service (YouTube) for legitimate content retrieval.

mmk-youtube-channel-summary