zoho-crm-connector

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references setup dependencies and configuration files hosted on the vendor's official GitHub repository and marketplace domains.
  • [COMMAND_EXECUTION]: Provides standard instructions for installing required integration components using platform-native commands.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface as it is designed to ingest data from Zoho CRM and perform record modifications. This is inherent to its primary function as a CRM connector.
  • Ingestion points: Data is retrieved from Zoho CRM modules via the list_records, get_record, search_records, and query_records actions.
  • Boundary markers: No specific delimiters or "ignore" instructions for ingested content are present in the skill body.
  • Capability inventory: The skill includes write capabilities through create_records, update_records, and delete_records actions.
  • Sanitization: Sanitization and validation are handled by the AgentPMT-hosted tool execution environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 03:14 AM