skills/deepgram/wsh/tui/Gen Agent Trust Hub

tui

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for the agent to execute shell commands to bootstrap its environment, specifically starting a local server (wsh server -L agent-$$ --ephemeral &) and managing process lifecycle (kill %1) during TUI sessions.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it requires the agent to read and act upon potentially untrusted text displayed within terminal applications.
  • Ingestion points: Terminal screen grid data and status messages retrieved via wsh_get_screen or interpreted from the 2D layout (SKILL.md).
  • Boundary markers: None; the skill does not instruct the agent to use delimiters or ignore instructions that might be embedded in the terminal output.
  • Capability inventory: The agent can perform keyboard input injection (wsh_send_input), manage background processes (wsh server), and interact with local unix sockets via curl (SKILL.md).
  • Sanitization: There are no instructions for sanitizing or escaping the content read from the terminal screen before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 04:12 AM
Security Audit — agent-trust-hub — tui