The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided descriptions of computational workloads to generate integration plans and performance checklists. It lacks explicit boundary markers or instructions to sanitize user input, creating a surface where malicious instructions embedded in technical requirements could influence the agent's output.
Ingestion points: User queries regarding "high-performance frontend computation" and "WASM vs IPC trade-offs" identified in the 'When to use this skill' section of SKILL.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded instructions within user data are present.
Capability inventory: The skill provides examples in usage.md using Tauri's invoke and emit APIs, and includes a capability definition in capabilities.json with core:default permissions.
Sanitization: No input sanitization or validation logic is defined for the requirement analysis phase.
[EXTERNAL_DOWNLOADS]: The skill references external third-party resources for implementation guidance and code dependencies.
Evidence: SKILL.md contains links to a community GitHub repository (p1mo/tauri-wasm) and mentions the tauri-wasm crate. It also links to technical articles on platforms such as Zhihu and CSDN. These resources are community-maintained rather than originating from official framework or trusted vendor repositories.

tauri-app-wasm