b2c-ecdn
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for the agent to perform management tasks (cache purging, firewall configuration, certificate updates) by executing shell commands with the
b2cCLI tool. - [EXTERNAL_DOWNLOADS]: The skill references the
@salesforce/b2c-clipackage, which is downloaded from a well-known package registry. This tool originates from a recognized technology provider. - [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection. It processes user-supplied data—such as domain names, cache tags, and file paths—and interpolates them directly into shell commands. There are no instructions for sanitizing these inputs or using boundary markers to prevent command or argument injection.
- Ingestion points: User queries supplying values for flags like
--domain-name,--path,--tag, and--hostnameinSKILL.md,references/ADVANCED.md, andreferences/SECURITY.md. - Boundary markers: Absent.
- Capability inventory: Shell command execution via the
b2cCLI inSKILL.md,references/ADVANCED.md, andreferences/SECURITY.md. - Sanitization: No validation or escaping logic is defined for user-provided parameters.
Audit Metadata