yt-script
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from external sources and uses it in a workflow that includes local script execution.\n
- Ingestion points: The skill reads a voice profile from the vendor-specific path
~/.uplika/<workspace-slug>/voice.mdand retrieves data from thesearch_youtube_videostool.\n - Boundary markers: There are no explicit instructions or delimiters used to prevent the agent from following instructions potentially embedded in the voice profile or search results.\n
- Capability inventory: The skill utilizes shell execution to run the
scripts/hookscore.pyscript.\n - Sanitization: No sanitization or validation of the ingested external data is performed before it is used to generate the hooks that are then passed to the scoring script.\n- [COMMAND_EXECUTION]: The skill relies on the execution of a local Python script,
scripts/hookscore.py, to perform its primary function of scoring and ranking YouTube hooks. While the script is part of the skill's own package, this represents a reliance on shell capabilities to process agent-generated content.
Audit Metadata