saleor
Pass
Audited by Gen Agent Trust Hub on Apr 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Installs the vendor-provided CLI tool '@membranehq/cli' from the official NPM registry to facilitate platform interactions.\n- [COMMAND_EXECUTION]: Utilizes system commands via the 'membrane' CLI to manage authentication, establish connections, and perform data operations.\n- [DATA_EXFILTRATION]: Facilitates the transfer of e-commerce data (orders, customers, products) between the local agent environment, the Saleor API, and the Membrane cloud platform.\n- [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it processes data from external Saleor instances and accepts natural language intent for action discovery and creation.\n
- Ingestion points: Data retrieved from Saleor API actions (products, orders, customers) and user-provided strings in '--intent' or 'action create' commands.\n
- Boundary markers: None explicitly defined in the instruction set.\n
- Capability inventory: Execution of shell commands via CLI, remote package installation, and dynamic creation of integration actions.\n
- Sanitization: Validation and sanitization are managed by the Membrane platform's backend services.
Audit Metadata