bat-submit

Pass

Audited by Gen Agent Trust Hub on Jul 6, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection. Ingestion points: The skill actively crawls external websites provided by the user as part of the extraction process (referencing references/01-extract.md). Boundary markers: There are no explicit delimiters or 'ignore' instructions wrapping the ingested content during prompt interpolation. Capability inventory: The skill utilizes bat-cli for network submission and local file writing, and runs a Python script (scripts/extract_video.py) for HTML parsing. Sanitization: The skill lacks automated sanitization or filtering of the content retrieved from external sources.
  • [COMMAND_EXECUTION]: The skill requires the execution of a local Python utility (scripts/extract_video.py) to extract media links from HTML. It also relies on the @bataitools/bat-cli command-line tool for authentication and data submission. These actions are within the expected scope of the skill's functionality.
  • [EXTERNAL_DOWNLOADS]: The installation of the @bataitools/bat-cli package from the NPM registry is required. As this package belongs to the skill's authoring organization ('bataitools'), it is considered a legitimate dependency for the tool's operation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 6, 2026, 01:43 AM
Security Audit — agent-trust-hub — bat-submit