firestore-operations-manager

Pass

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • Indirect Prompt Injection (LOW): The skill processes data from Firestore documents, creating a vulnerability surface where instructions embedded in the database could influence agent behavior.
  • Ingestion points: Firestore collections (specifically agent_memory and public documents) read during CRUD operations.
  • Boundary markers: Absent; there are no instructions to the agent to treat retrieved database content as untrusted or to use delimiters.
  • Capability inventory: Bash(cmd:*), Write, Edit, Read. The high-privilege bash access increases the impact of a successful injection.
  • Sanitization: Absent; the skill does not specify validation or sanitization steps for data retrieved from Firestore.
  • Command Execution (SAFE): The skill requires Bash(cmd:*) to execute standard infrastructure tools like gcloud and firebase-tools. This is consistent with the skill's stated purpose of managing Firestore operations and setup.
  • Metadata Analysis (SAFE): The metadata provided (author, description, and version) is consistent with the provided scripts and documentation. No deceptive patterns were found in the metadata fields.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 18, 2026, 09:18 PM