The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill implements patterns for ingesting and processing data from external, potentially untrusted sources like Cloud Firestore and Generative AI models.
Ingestion points: Data is pulled into the application via Firestore observables (references/product-firestore.md), Realtime Database listeners (references/product-realtime-database.md), and generative responses from Gemini models (references/product-vertexai.md).
Boundary markers: The provided code snippets do not include explicit prompt delimiters or instructions to the agent to ignore embedded commands within the fetched data.
Capability inventory: The skill manages network communication with Firebase APIs, performs database read/write operations, and handles authentication state.
Sanitization: The implementation examples do not show explicit sanitization or validation of data before it is processed or displayed.
[COMMAND_EXECUTION]: The skill documents the use of various CLI tools for project setup, testing, and deployment.
Evidence: Instructions include running ng add @angular/fire, ng deploy, and firebase emulators:start to manage the development lifecycle.
[EXTERNAL_DOWNLOADS]: The skill utilizes external dependencies and script references from well-known providers.
Evidence: The service worker configuration (references/product-messaging.md) imports Firebase SDK libraries from Google's gstatic.com domain. It also defines dependencies on official Angular and Firebase packages via NPM.

angularfire