firebase-data-connect

Warn

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [Command Execution] (MEDIUM): The skill includes instructions and tools for running Firebase CLI commands such as firebase init and firebase deploy. These operations perform file system writes and cloud deployments, which are high-impact actions.\n- [Indirect Prompt Injection] (MEDIUM): The workflow for generating backend schemas and queries from user requirements is susceptible to indirect prompt injection.\n
  • Ingestion points: Schema and query definitions (schema.gql, queries.gql) derived from user prompts.\n
  • Boundary markers: No delimiters or safety instructions are used to prevent the agent from obeying instructions embedded in the user's data model requirements.\n
  • Capability inventory: firebase deploy and firebase_init provide the ability to modify project state and deploy code.\n
  • Sanitization: The skill does not include steps to sanitize or validate the generated GraphQL schema or authorization logic before deployment.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 07:44 AM