gladia-documentation-auto
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill provides installation instructions for the Gladia CLI using a piped-to-shell pattern (e.g.,
curl ... | shandpowershell ... | iex). These commands download and execute setup scripts from the vendor's official GitHub repository (gladiaio/gladia-cli). This is a documented, standard procedure for tool installation from a known vendor. - [EXTERNAL_DOWNLOADS]: The documentation references external packages for the Gladia SDK on NPM (
@gladiaio/sdk) and PyPI (gladiaio-sdk). These are official vendor libraries used for interacting with the service. - [COMMAND_EXECUTION]: Example commands are provided for the
gladiaCLI tool to perform authentication and audio transcription tasks. These are intended for user or agent use after the tool is installed. - [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: The skill facilitates processing audio and video files from external URLs or local paths through the Gladia API endpoints.
- Boundary markers: No specific boundary markers or instruction-ignoring delimiters are defined for handling the resulting transcripts.
- Capability inventory: The skill allows for network requests to the Gladia API and local file uploads for transcription.
- Sanitization: The skill relies on the upstream API for processing and does not specify internal sanitization for returned transcript content.
Audit Metadata