scenario-admin-analytics
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill demonstrates high security maturity by implementing proactive defenses. The Python helper script (
scripts/analytics.py) includes a dedicated check (FORMULA_PREFIXES) to prevent CSV Formula Injection attacks. Additionally, the HTML dashboard inassets/dashboard.htmldefines a restrictive Content Security Policy (CSP) that prevents unauthorized script execution or data exfiltration. - [SAFE]: Data handling is performed using restrictive file system permissions. The script explicitly sets directories to
0o700and sensitive PDF reports to0o600to ensure local data is not readable by other users on the system. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted usage data (user names, model IDs) retrieved from the Scenario MCP. While this presents a surface for indirect prompt injection, the skill mitigates this through robust sanitization, including HTML escaping for the dashboard and data validation in the Python processing script.
- [COMMAND_EXECUTION]: The skill uses a local Python script (
scripts/analytics.py) for data transformation. This script is bundled with the skill, performs no network operations, and relies exclusively on standard libraries, minimizing the risk of unauthorized execution.
Audit Metadata