The Agent Skills Directory

[COMMAND_EXECUTION] (HIGH): The skill executes a Python script via uv run and passes user-provided arguments (app-name, display-name, bundle-id) directly into the shell command. A malicious user could provide a string like my-app\"; touch /tmp/pwned; # to escape the command context and execute arbitrary code on the system.
[PROMPT_INJECTION] (HIGH): The skill is highly susceptible to Indirect Prompt Injection (Category 8). It ingests untrusted user data (App details) and uses it to drive high-privilege operations including Bash commands and Write operations. Evidence: (1) Ingestion points: User input in Step 1. (2) Boundary markers: Absent. (3) Capability inventory: Bash tool usage and file system writing. (4) Sanitization: None specified in the instructions.
[EXTERNAL_DOWNLOADS] (LOW): The skill instructions lead to the execution of pnpm install, which downloads and installs numerous third-party dependencies from the npm registry. While expected for scaffolding, it represents a standard supply chain risk.

tauri-scaffold