arete
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill workflow involves ingesting data from external discovery commands and using the results to parameterize subsequent installation and management commands.
- Ingestion points: Untrusted data enters the context through the output of discovery commands like
a4 know searchanda4 exploreinSKILL.md. - Boundary markers: Absent; the instructions prompt the agent to use returned stack and program references directly in installation commands without explicit delimiters or validation steps.
- Capability inventory: The skill utilizes shell command execution for installation (
a4 install), updates (a4 update), and configuration management (a4 config). - Sanitization: The skill relies on the internal validation logic of the
a4CLI tool to sanitize descriptors before execution.
Audit Metadata