The Agent Skills Directory

[SAFE]: The skill provides robust security patterns for Electron development, including the correct configuration of BrowserWindow with contextIsolation: true, nodeIntegration: false, and sandbox: true to mitigate exploitation risks.
[SAFE]: It implements secure Inter-Process Communication (IPC) using contextBridge with explicit channel whitelisting and includes a main-process helper to verify the origin and frame of incoming IPC messages.
[SAFE]: The guide includes advanced security measures such as Content Security Policy (CSP) implementation, navigation hijacking prevention, and a custom protocol handler with built-in protection against path traversal attacks.
[INDIRECT_PROMPT_INJECTION]: The skill involves analyzing external project files and code, which is a standard surface for indirect prompt injection.
Ingestion points: Project structure and source code files as described in the instructions of references/detailed-guide.md.
Boundary markers: The skill emphasizes application-level boundary enforcement through IPC whitelisting and input validation.
Capability inventory: The agent typically uses file read/write and shell execution capabilities to support the development processes.
Sanitization: The skill provides templates for sanitizing inputs within the Electron app, which the agent follows when generating code.

electron-development