suggest-sales-next-step

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • Integrated Data Processing: The skill aggregates data from external sources like meeting notes and message threads. As a standard practice, it is designed to use this information to inform its suggestions while maintaining the specific source boundaries defined in its configuration.
  • External Data Context: The skill is designed to ingest data from multiple sources to provide context-aware recommendations. Ingestion points: Data enters via the calendar, crm, meeting_notes, document_store, and messaging connectors. Boundary markers: The instructions do not explicitly define delimiters for external content within the prompts. Capability inventory: The skill uses MCP tools for data retrieval and interacts with other internal components like sales:user-context. Sanitization: No specific sanitization or filtering steps for retrieved content are described in the instructions.
  • Automated Workflow Management: It supports scheduled execution through a 'heartbeat' mechanism, which allows for periodic sales check-ins. This functionality is governed by user-defined preferences for notifications and quiet periods.
  • State Persistence: User onboarding and experience levels are tracked via a local state file ($CODEX_HOME/state/plugins/role-specific-plugins/sales/onboarding-state.json). This allows the skill to provide a consistent experience by recognizing previously tried workflows.
  • Structured Skill Chaining: The skill invokes a secondary configuration component (sales:user-context) to resolve environment-specific variables and source permissions before performing any audits.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 06:13 AM
Security Audit — agent-trust-hub — suggest-sales-next-step