the-trade-desk
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
membraneCLI tool for managing connections and executing actions (e.g.,membrane action run). These commands are run in the local environment to interface with The Trade Desk data services. - [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the
@membranehq/clipackage from the NPM registry. This is an official tool provided by the vendor for interacting with their platform. - [PROMPT_INJECTION]: The skill identifies a potential surface for indirect prompt injection (Category 8) due to the handling of user-supplied data in action queries. Ingestion points: User-provided strings are interpolated into the
--intentand--inputparameters of themembraneCLI (SKILL.md). Boundary markers: No explicit delimiters or instructions are used to separate untrusted user input from the command logic. Capability inventory: The CLI allows the agent to read and modify data on The Trade Desk platform based on the results of the intent-based search. Sanitization: Input validation or escaping mechanisms for user-provided intents are not specified in the instructions.
Audit Metadata