The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill includes instructions to read and process data from Cloud Firestore documents, which serves as a potential surface for indirect prompt injection if the stored data is untrusted.
Ingestion points: The querySnapshot.docs and doc.data() calls in SKILL.md bring external database content into the application context.
Boundary markers: The skill does not explicitly use delimiters for the retrieved data in the provided logic snippets.
Capability inventory: The skill utilizes network operations for Firestore synchronization and write capabilities via set, add, and batch operations.
Sanitization: The skill proactively recommends implementing Firebase Security Rules and validating user input before submission to mitigate risks associated with untrusted data.
[EXTERNAL_DOWNLOADS]: The skill references an external Git repository to optimize build times for iOS and macOS platforms.
Evidence: It suggests using a pre-compiled framework from https://github.com/invertase/firestore-ios-sdk-frameworks.git, which is the official repository for FlutterFire's pre-compiled Firestore SDKs maintained by Invertase.

firebase-cloud-firestore