The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill defines a template-based workflow in templates/basic.md that ingests user-provided values into the agent's execution context.
Ingestion points: Variables such as {project_name}, {frontend}, {pm}, and {dev_url} are populated by user input and used to construct commands.
Boundary markers: The templates and instructions do not include boundary markers or delimiters to isolate untrusted user data from the generated instructions.
Capability inventory: SKILL.md instructs the agent to run creation commands, install dependencies, and initiate dev mode, all of which typically involve shell command execution.
Sanitization: There is no evidence of validation or sanitization requirements for these inputs before they are interpolated into executable shell strings, potentially allowing command injection if malicious inputs are provided.
[EXTERNAL_DOWNLOADS]: The skill references official Tauri v2 installation methods and documentation.
Evidence: SKILL.md points to https://v2.tauri.app/ for project setup. These references include standard installation patterns such as curl pipe-to-shell for the create-tauri-app utility, which are official distribution channels for the framework.

tauri-app-creator