tauri-mcp-cli

Pass

Audited by Gen Agent Trust Hub on Mar 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONREMOTE_CODE_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the tauri-mcp CLI tool to manage Tauri driver sessions, capture UI state, and perform automation tasks via terminal commands.
  • [REMOTE_CODE_EXECUTION]: The tauri-mcp webview-execute-js command enables the dynamic execution of arbitrary JavaScript scripts within the application's webview context.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface as it ingests untrusted data from the application and provides tools to execute actions based on that data. 1. Ingestion points: The agent can read potentially attacker-controlled data via tauri-mcp read-logs, tauri-mcp webview-find-element, and tauri-mcp ipc-get-captured (SKILL.md). 2. Boundary markers: There are no instructions or delimiters provided to distinguish untrusted application data from the agent's core instructions. 3. Capability inventory: The skill provides powerful interaction capabilities, including tauri-mcp webview-execute-js, tauri-mcp ipc-execute-command, and various UI interaction tools (SKILL.md). 4. Sanitization: No sanitization, escaping, or validation of the data retrieved from the application is described before it is used in subsequent automation commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 16, 2026, 12:47 PM