agy-native
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the 'agy' command-line utility to execute packets and manage sessions. It instructs the agent to dynamically discover the tool's capabilities using 'agy --help'. It also documents flags like '--dangerously-skip-permissions' for high-privilege execution, while recommending '--sandbox' for security.
- [PROMPT_INJECTION]: The skill processes external data referred to as 'packets' and 'validator contexts', which creates a surface for indirect prompt injection. 1. Ingestion points: 'explicit-packet' and fresh validator contexts in SKILL.md. 2. Boundary markers: None explicitly mentioned to separate data from instructions. 3. Capability inventory: Execution of packets and tool invocation via the 'agy' CLI. 4. Sanitization: No explicit sanitization or filtering of packet content is described.
Audit Metadata