desktop-tauri
Warn
Audited by Gen Agent Trust Hub on Jul 8, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The configuration snippet in SKILL.md references a JSON schema hosted on the 'nicedoc' GitHub organization rather than the official Tauri project repositories. This repository is not verified or part of the trusted vendor list.\n- [COMMAND_EXECUTION]: Eight reference files (including 'architecture-patterns.md', 'code-organization.md', and 'security-best-practices.md') contain 150 repetitive technical blocks each, totaling approximately 1,200 redundant sections. This pattern of technical 'chaff' is a signature technique used to obfuscate malicious content or exhaust the context window and token limits of automated security analysis systems.\n- [DATA_EXFILTRATION]: The skill provides implementation templates that combine local file system access (via tauri-plugin-fs) with outbound network request capabilities (via tauri-plugin-http or the Rust reqwest crate). While functional for Tauri apps, providing these capabilities in tandem without explicit security guards or boundary warnings creates an architectural path for sensitive data exfiltration.\n- [PROMPT_INJECTION]: The skill lacks sufficient boundary markers and sanitization logic when teaching patterns that ingest external data (such as reading from the file system or fetching from URLs) while simultaneously exposing powerful shell execution and native API capabilities to the frontend WebView renderer.
Audit Metadata