recurring-reports-post-offboarding
Pass
Audited by Gen Agent Trust Hub on Apr 9, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface as it is designed to ingest and process data from external, untrusted sources without explicit security boundaries.
- Ingestion points: Support tickets, Slack threads (accessed via Slack MCP), and Jira tickets (accessed via Atlassian MCP).
- Boundary markers: Absent. There are no instructions to the agent to ignore or delimit potentially malicious instructions within the ingested data.
- Capability inventory: The skill utilizes BigQuery SQL execution (
execute_sql) and provides Python code snippets for execution in a Django shell environment. - Sanitization: Absent. The provided queries and scripts interpolate user-supplied data, such as
{wsid}and{recipient_email}, directly into the execution strings without validation or escaping. - [COMMAND_EXECUTION]: The skill provides Python and SQL snippets that incorporate variables derived from external inputs. If an attacker influences these inputs via a support ticket or Slack message, they could attempt SQL or Python code injection to gain unauthorized access to data or manipulate the application environment.
Audit Metadata