datarobot
Pass
Audited by Gen Agent Trust Hub on Mar 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the
@membranehq/cliNode.js package. This is a vendor-provided tool used to facilitate communication between the agent and the DataRobot API via the Membrane platform. - [COMMAND_EXECUTION]: The skill instructions involve executing shell commands using the
membraneCLI to manage authentication, discover actions, and interact with DataRobot resources. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes data from an external source (DataRobot).
- Ingestion points: Data is brought into the agent context through the output of
membrane action runandmembrane requestcommands. - Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat data from these commands as untrusted.
- Capability inventory: The agent can execute arbitrary shell commands and network requests via the
membraneCLI as described in SKILL.md. - Sanitization: No sanitization or validation of the data returned by the DataRobot API is performed before it is presented to the agent.
Audit Metadata