zoho-crm
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection as it is instructed to fetch and process data from CRM records (Ingestion point:
zoho.crm.v8.getRecordsreferenced inreferences/functions.md). - Boundary markers are absent in the logic for processing record field values.
- The skill has capabilities to perform write operations to the CRM and make external network requests (
invokeUrl). - No explicit sanitization or filtering logic is specified for data values before they are used in prompts.
- [EXTERNAL_DOWNLOADS]: The skill fetches OpenAPI specifications from Zoho's official GitHub repository (
github.com/zoho/crm-oas.git) to resolve API signatures (referenced inreferences/api.md). This is a legitimate operation from a trusted source. - [DYNAMIC_EXECUTION]: The skill utilizes an included Python script (
extract_schema.py) to parse downloaded API documentation at runtime (referenced inreferences/api.md). This execution is part of the skill's intended functionality to ensure code accuracy.
Audit Metadata