motherduck-cli
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill provides commands to install the CLI by piping remote scripts from the vendor's domain ('install.motherduck.com') directly into shell and PowerShell interpreters. This is the documented installation path for the tool.\n- [EXTERNAL_DOWNLOADS]: The CLI installation process involves fetching setup scripts from an external vendor-controlled domain ('install.motherduck.com').\n- [COMMAND_EXECUTION]: The skill's primary function is to execute various 'motherduck' shell commands for database queries, project management, and automated workflows.\n- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and process output from CLI commands, such as query results and logs, which can contain data from untrusted external database records or execution outputs.\n
- Ingestion points: CLI output from 'motherduck query', JSON/NDJSON log files from 'motherduck dive watch', and log streams from 'motherduck flight logs'.\n
- Boundary markers: The skill advises using machine-readable JSON/CSV formats and filtering output with 'jq' to limit the data brought into context.\n
- Capability inventory: The agent has the ability to execute a full range of 'motherduck' CLI commands to manipulate remote database resources.\n
- Sanitization: While structured output and filtering are recommended, there are no specific instructions to escape or sanitize data values before they are processed by the agent.
Audit Metadata