firebase-firestore

Pass

Audited by Gen Agent Trust Hub on Nov 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the official firebase-tools CLI (via npx) to perform database operations such as listing instances, creating new databases, and deploying security rules or indexes. These are standard administrative tasks for Firebase developers.
  • [EXTERNAL_DOWNLOADS]: The instructions include commands to fetch the latest SDK version numbers from Google's official Maven repository (dl.google.com). This ensures the agent uses up-to-date dependencies rather than hardcoded, potentially vulnerable versions.
  • [DYNAMIC_EXECUTION]: In the Enterprise edition guides, the agent is instructed to read documentation files extracted from the source JARs of the Firestore SDK within the local build cache. This allows the agent to access version-specific documentation for advanced features like Query Pipelines without relying on external or untrusted websites.
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of database records and the creation of configuration files. While these are common attack surfaces, the skill does not contain instructions that would lead to the automatic execution of data as code, and the use of the CLI provides a natural layer of user oversight.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 3, 2026, 09:18 PM