tmux
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
tmux send-keyscommand to inject keystrokes and text directly into terminal sessions, which allows the agent to execute shell commands or interact with terminal applications. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from terminal panes which may contain untrusted content (e.g., logs from a web server or output from an external process).
- Ingestion points: Terminal content is ingested via
tmux capture-paneinSKILL.mdand thescripts/wait-for-text.shscript. - Boundary markers: The skill does not implement delimiters or explicit 'ignore instructions' warnings when reading output from the terminal.
- Capability inventory: The agent possesses the capability to send keystrokes back to the terminal (
tmux send-keys) and has access to standard command execution tools. - Sanitization: The scripts do not perform sanitization, filtering, or validation on the captured terminal text before it enters the agent context.
Audit Metadata