crm-sync
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes bundled Python scripts from the plugin's root directory to perform CRM interactions, check status, and manage the approval lifecycle. The commands use interpolated placeholders for brand identifiers and data payloads, which is the intended mechanism for this tool's operation.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from CSV and JSON files, creating an attack surface for indirect instructions.
- Ingestion points: User-provided CSV files, JSON arrays, and data retrieved from other connected platforms (e.g., Google Sheets, email platforms) as described in SKILL.md.
- Boundary markers: Not explicitly defined in the prompt instructions for data interpolation.
- Capability inventory: The skill possesses file-reading capabilities, shell command execution (via bundled Python scripts), and network communication through connected CRM MCPs.
- Sanitization: The process includes data validation for specific formats, including RFC 5322 for emails and E.164 for phone numbers, along with picklist value verification.
Audit Metadata