zentty-cli
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted data from the terminal environment and session history, creating a surface for indirect prompt injection attacks.\n
- Ingestion points: Untrusted data enters the agent's context through the
history search,history show, andpane readcommands defined inSKILL.md.\n - Boundary markers: The skill provides explicit instructions to the agent to treat content from session history as data rather than instructions, noting that a
contentIsUntrustedflag is provided to signal risks.\n - Capability inventory: The skill grants the agent capabilities to influence other processes, including sending text or keys to terminal panes (
pane send,pane send-key) and starting new processes (launch,float new,overlay open).\n - Sanitization: No programmatic sanitization or validation of the ingested text is specified beyond the behavioral instructions to the agent.
Audit Metadata