skills/spotdraft/oncall-investigator-skills/clickwrap-tray-salesforce-sync-triage/Gen Agent Trust Hub
clickwrap-tray-salesforce-sync-triage
Pass
Audited by Gen Agent Trust Hub on Apr 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions to execute management commands (e.g.,
python manage.py retrigger_kp_created_webhook_for_clickthrough_contracts) that interpolate user-provided variables such as{workspace_id},{start_date}, and{end_date}. This pattern creates a risk of command injection if the inputs are not properly validated by the platform. - [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as defined in Category 8.
- Ingestion points: Untrusted data enters the agent context through parameters like
{contract_id},{workspace_id}, and date strings inSKILL.md. - Boundary markers: None are present; user inputs are directly concatenated into SQL templates and shell commands.
- Capability inventory: The skill utilizes BigQuery for database lookups and shell execution via the Django management CLI.
- Sanitization: No validation or escaping mechanisms are described within the skill for the processed inputs.
Audit Metadata