periodic-sales-performance-review

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists of high-level instructions and configuration templates for sales analysis. It does not include executable scripts or commands that pose a risk to the system. No malicious patterns were identified in the metadata or the instructional content.
  • [PROMPT_INJECTION]: Analysis of the skill's operational flow identified an indirect prompt injection surface. The skill ingests untrusted data from CRM platforms and CSV files which is then processed into generated reports.
  • Ingestion points: Data is pulled in Step 1 from CRM APIs (Salesforce, HubSpot, etc.) and CSV files.
  • Boundary markers: The instructions do not define specific delimiters to isolate external data from the agent's logic.
  • Capability inventory: The skill performs data analysis and generates markdown-based reports and coaching recommendations.
  • Sanitization: There is no explicit requirement for the agent to sanitize or validate the external data strings before interpolation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 03:07 PM
Security Audit — agent-trust-hub — periodic-sales-performance-review