configurator-cli
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to use
pnpm dlxandnpxto fetch and execute the@saleor/configuratorpackage from the NPM registry. These are standard procedures for running the official CLI tool provided by the vendor. - [COMMAND_EXECUTION]: The skill documents several CLI commands (
validate,introspect,deploy,diff) that perform network operations to synchronize local configuration with Saleor GraphQL endpoints and file system operations to read/write configuration files (config.yml) and deployment reports. These actions are aligned with the skill's primary purpose of store configuration management. - [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it processes external data that could potentially contain malicious instructions.
- Ingestion points: The agent ingests data from local
config.ymlfiles and remote data fetched from Saleor store instances via theintrospectanddiffcommands. - Boundary markers: While the CLI supports structured JSON output which helps delimit data, there are no explicit boundary markers or instructions to ignore potential natural language commands embedded within the YAML configuration content.
- Capability inventory: The skill possesses capabilities to perform network requests and write to the local file system (e.g., updating configuration files or generating reports).
- Sanitization: No specific sanitization or validation mechanisms are described to filter out or escape natural language instructions within the processed configuration data.
Audit Metadata