youtube-transcript-extract
Pass
Audited by Gen Agent Trust Hub on Apr 19, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches the
ytsubspackage from the public npm registry using thenpxcommand to perform its primary function. - [REMOTE_CODE_EXECUTION]: Executes the
ytsubsutility package downloaded from the npm registry vianpx -yon the local system. - [COMMAND_EXECUTION]: Orchestrates the extraction of YouTube transcripts by running shell commands with user-provided video URLs or IDs.
- [PROMPT_INJECTION]: The skill processes external data (YouTube transcripts) which represents a surface for indirect prompt injection. Ingestion points: the output of the
ytsubscommand (transcripts and metadata) is ingested into the agent's context. Boundary markers: none are specified in the instructions or sample output to delimit external content from instructions. Capability inventory: the skill usesnpxfor command execution and file system access for saving extracted transcripts. Sanitization: no explicit sanitization or filtering of the retrieved transcript content is performed before it is processed by the agent.
Audit Metadata