skills/spotdraft/oncall-investigator-skills/approval-v5-bulk-reassign-download-restriction-backend/Gen Agent Trust Hub
approval-v5-bulk-reassign-download-restriction-backend
Pass
Audited by Gen Agent Trust Hub on Apr 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for executing SQL queries in BigQuery and running data modification logic within a production Django shell environment to perform bulk updates on approval records.\n- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) where user-supplied identifiers are interpolated into database queries.\n
- Ingestion points: Variable placeholders such as
{workspace_id},{contract_id}, and{approval_name_1}inSKILL.md.\n - Boundary markers: Absent; user inputs are directly placed into SQL string literals without delimiters.\n
- Capability inventory: The skill possesses the capability to query production databases via BigQuery and modify production state through the Django shell.\n
- Sanitization: No input validation, escaping, or sanitization logic is described for the user-provided parameters.\n- [DATA_EXFILTRATION]: The skill facilitates the retrieval of sensitive contract and approval information from production environments. While this is the documented purpose for support workflows, the lack of input sanitization could be leveraged to access unauthorized data records.
Audit Metadata