shopware-cli-extension-anonymize
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local CLI commands using the
shopware-cliutility, a tool provided by the vendor, to initialize and validate extension configurations.\n - Evidence: Instructs the agent to run
shopware-cli extension config init,shopware-cli extension config-schema, andshopware-cli extension validate ..\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted source code and metadata from the extension directory to identify configuration keys, creating a surface for indirect prompt injection.\n - Ingestion points: Reads extension files including
composer.json,manifest.xml, PHP entity definitions, and SQL migrations to find table names, column names, and configuration keys.\n - Boundary markers: Lacks specific instructions for the agent to wrap extracted data in delimiters or treat it as untrusted input during the interpolation process.\n
- Capability inventory: The skill provides the agent with the ability to modify local configuration files (
.shopware-extension.yml) and execute local shell commands via theshopware-clitool.\n - Sanitization: Includes a defensive instruction directing the agent not to print live secret values from config,
.env, or a database in its summary output to the developer.
Audit Metadata