coupontools
Pass
Audited by Gen Agent Trust Hub on Nov 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from the Coupontools platform, which represents an untrusted external source. \n
- Ingestion points: Data returned from
membrane action list,membrane action run, andmembrane request(e.g., coupon details, user directories, and session information) is entered into the agent's context. \n - Boundary markers: No specific delimiters or instructions to ignore embedded instructions within the retrieved data are provided to the agent. \n
- Capability inventory: The skill allows the agent to execute shell commands via the
membraneCLI to perform sensitive actions like creating users, sending emails/SMS, and updating coupon status. \n - Sanitization: The instructions do not specify any sanitization, validation, or escaping of the external content before it is processed by the agent.
- [EXTERNAL_DOWNLOADS]: The skill instructs the installation of the Membrane CLI tool from a public package registry. \n
- Evidence:
npm install -g @membranehq/cli@latest. - [COMMAND_EXECUTION]: The skill uses the
membraneCLI to interact with the Coupontools API and manage service connections.
Audit Metadata