media-notes-publishing
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources that are not under the direct control of the user or the skill author.
- Ingestion points: The
scripts/acquire_subtitle.pyscript takes a user-provided<URL>(e.g., from Bilibili or YouTube) as input to fetch subtitles. - Boundary markers: The skill does not define explicit delimiters or instructions to the agent to ignore embedded commands within the fetched subtitle content.
- Capability inventory: The skill includes capabilities to execute shell commands via
pythonandnodeand perform file writes to the local filesystem (media-note/and the site repository). - Sanitization: There is no evidence of sanitization or filtering of the extracted subtitle text before it is presented to the agent for note generation.
- [COMMAND_EXECUTION]: The skill instructions direct the agent to execute local scripts and commands to perform its tasks.
- Evidence: Execution of
python <skill>/scripts/acquire_subtitle.pyandnode scripts/publish-notes.mjs. These scripts interact with the local filesystem and external network resources to fetch media content and publish files to a target directory.
Audit Metadata