skills/zoho/crm-skills/zoho-crm/Gen Agent Trust Hub

zoho-crm

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the generation and validation of CRM automation scripts which may process untrusted data from record fields.
  • Ingestion points: Data from CRM modules (Leads, Contacts, etc.) and API responses are processed by the agent during the 5-step reasoning flow in SKILL.md.
  • Boundary markers: Absent. The instructions do not define delimiters or provide warnings to disregard instructions embedded within CRM data being processed.
  • Capability inventory: The skill generates code capable of performing network requests (invokeUrl, zrc.get), database writes (ZDK.Apps.CRM), and sending emails (send-mail).
  • Sanitization: No explicit guidance is provided for sanitizing or escaping CRM field content before it is used in code generation or validated by the agent.
  • [DYNAMIC_EXECUTION]: The skill is designed to generate executable Deluge and ClientScript code for CRM automation.
  • This functionality is the primary purpose of the skill and is consistent with the 'Zoho CRM Orchestration' use case.
  • [COMMAND_EXECUTION]: The skill includes a Python utility script (assets/api/extract_schema.py) for processing OpenAPI specifications.
  • The script uses standard libraries and the pyyaml package to perform file I/O within the skill's internal directory structure.
  • [EXTERNAL_DOWNLOADS]: The skill references official vendor domains (zoho.com, zohocrm.dev, zohoapis.com) and the author's GitHub repository in its documentation and README.
  • These references point to authoritative API documentation and official installation endpoints.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 02:28 PM
Security Audit — agent-trust-hub — zoho-crm