The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill uses a template file (templates/basic.md) containing placeholders that represent a potential surface for indirect prompt injection if external or untrusted data is interpolated without validation.
Ingestion points: The file templates/basic.md includes multiple placeholders such as {product_name}, {version}, {dev_url}, {frontend_dist}, and {title} which are intended to be filled with external project-specific data.
Boundary markers: There are no explicit boundary markers or instructions to the agent to treat interpolated content as untrusted or to ignore any embedded commands within these fields.
Capability inventory: The skill facilitates the generation and modification of the tauri.conf.json file, which controls critical application security settings like Content Security Policy (CSP) and platform-specific permissions.
Sanitization: The skill does not provide any mechanisms or instructions for sanitizing or validating the inputs before they are placed into the template.

tauri-config