QuickBooks Automation

Pass

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: SAFE
Full Analysis
  • [Data Exposure & Exfiltration] (SAFE): The skill interacts with QuickBooks Online through the specified Rube MCP server and Composio toolkit. No exfiltration to unauthorized or suspicious third-party domains was observed.
  • [Prompt Injection] (SAFE): The skill instructions are purely functional and do not contain patterns attempting to override agent behavior, bypass safety filters, or extract system prompts.
  • [Indirect Prompt Injection] (LOW): As a data-processing skill, it ingests external records (customers, invoices) that could potentially contain malicious instructions. However, this is an inherent risk of the integration type rather than a specific malicious design in the skill itself.
  • Ingestion points: Data is retrieved via QUICKBOOKS_READ_CUSTOMER, QUICKBOOKS_QUERY_ACCOUNT, and QUICKBOOKS_LIST_INVOICES.
  • Boundary markers: The skill does not explicitly define markers to delimit untrusted data.
  • Capability inventory: The skill possesses write capabilities such as QUICKBOOKS_CREATE_INVOICE and QUICKBOOKS_CREATE_CUSTOMER.
  • Sanitization: No specific sanitization logic is defined within the skill markdown.
  • [Dependencies] (SAFE): The dependency on the Rube MCP server is standard for this integration, and no unverifiable remote scripts or packages are downloaded or executed.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 19, 2026, 04:04 AM