firebase-firestore-standard

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides commands to deploy Firestore rules and indexes and start local emulators using the official firebase-tools CLI.
  • [EXTERNAL_DOWNLOADS]: The skill uses 'npx -y firebase-tools@latest', which downloads and executes the latest version of the official Firebase command-line utility from the NPM registry. This is a trusted tool from a well-known service.
  • [PROMPT_INJECTION]: The file 'references/security_rules.md' contains extensive instructions and a security-focused workflow for the agent to follow. These are task-specific instructions aimed at ensuring secure database configuration.
  • [PROMPT_INJECTION]: The skill instructs the agent to analyze the user's codebase (references/security_rules.md, Phase-1) to inform security rule generation. This exposes the agent to potential indirect prompt injection from content within the analyzed files.
  • Ingestion points: references/security_rules.md (Phase-1: Codebase Analysis).
  • Boundary markers: None provided.
  • Capability inventory: File system read (codebase scan) and file writing (firestore.rules).
  • Sanitization: Relies on the agent's reasoning and the mandatory 'Phase-3: Devil's Advocate' security validation workflow.
  • [SAFE]: All referenced tools, repositories, and domains (such as firebase.google.com) are official resources provided by the vendor, firebase.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 12:16 PM