The Agent Skills Directory

[PROMPT_INJECTION]: The skill facilitates the analysis of external, untrusted source code from Tauri projects, which presents a surface for indirect prompt injection attacks.
Ingestion points: The agent reads and processes user-provided files such as lib.rs, tauri.conf.json, and frontend source files.
Boundary markers: The instructions lack specific boundary markers or 'ignore embedded instruction' warnings to prevent the agent from being misled by malicious content within the analyzed files.
Capability inventory: The skill leverages host agent capabilities for file system access and text searching (read_file, grep).
Sanitization: No sanitization or filtering of the ingested file content is performed before processing.
[SAFE]: No hardcoded credentials, unauthorized network operations, or obfuscation techniques were detected in the skill's content.
[SAFE]: There is a metadata discrepancy between the provided author context (Impertio-Studio) and the author listed in the SKILL.md frontmatter (OpenAEC-Foundation). However, this does not appear to be deceptive or malicious.
[NO_CODE]: The skill is composed entirely of markdown documentation and guidance checklists, without any bundled scripts or executable code.

tauri-agents-review