tauri-debugger
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands for system diagnostics, including 'systemctl' to check service status and 'gdbus' to query desktop portal properties.
- [COMMAND_EXECUTION]: Suggests the use of 'sudo apt install' for installing system-level dependencies like 'zenity'.
- [REMOTE_CODE_EXECUTION]: References a script named 'scripts/webkit-dev.py' for testing purposes which is not included within the skill package, requiring the agent to execute unverified local files.
- [PROMPT_INJECTION]: Contains instructions to deliberately weaken security configurations by setting 'dangerousDisableAssetCspModification' and enabling 'unsafe-inline' in Content Security Policy (CSP) settings.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) as it is designed to ingest and act upon untrusted project data:
- Ingestion points: Reads project files including 'vite.config.ts', 'tauri.conf.json', 'Cargo.toml', and capability manifests.
- Boundary markers: None identified; the skill treats content from these files as structured data for analysis.
- Capability inventory: Uses 'systemctl', 'gdbus', 'cargo', and 'python3' based on the state of the analyzed files.
- Sanitization: No validation or sanitization of file content is described before use in diagnostics.
Audit Metadata