Enterprise Prompt Governance and System Instructions

1. Overview & Business Value#

This comprehensive operational guide details the architecture, business workflows, configuration parameters, and execution steps for Enterprise Prompt Governance and System Instructions within SmartERP.

Authoring enterprise system prompts, brand voice guidelines, Amharic localization directives, and anti-jailbreak controls.

{screenshot:prompt-governance-and-system-instructions-overview-dashboard}

2. Prerequisites & Security Access Groups#

• Required Modules: ashewa_ai_workforce, ashewa_ai_bridge, smarterp_ai_agents
• Security Groups:
  • AI Workforce Operator (Technical ID: ashewa_ai.group_ai_operator)
  • AI Systems Administrator (Technical ID: ashewa_ai.group_ai_admin)
  • Governance Officer (Technical ID: ashewa_ai.group_ai_governance)
• Master Data Dependencies: Autonomous AI agent profiles, n8n orchestrator webhooks, LLM model parameters, prompt governance rules, and human-in-the-loop approval queues.

3. Step-by-Step Configuration & Execution#

Step 1: Navigating to the Module Interface#

1. Log into your SmartERP instance at https://smarterp.et/odoo.
2. Open the main App Switcher (grid icon in the top-left corner).
3. Select the AI Workforce app icon.
4. From the top navigation bar, open AI Workforce → Copilots → Active Agents / Interaction History / Governance Queue.

{screenshot:prompt-governance-and-system-instructions-app-switcher-navigation}

Step 2: Creating and Parameterizing Records#

1. Click the New button in the top-left control panel.
2. Complete the mandatory fields in the record form:
   • Agent Name & Persona (Mandatory): Specialized AI role (Financial Auditor Copilot, SCM Copilot, etc.)
   • Target Operational Addon (Mandatory): Integrated SmartERP module (Accounting, HR, Inventory, Sales)
   • Autonomy Level (Mandatory): Level 1 Assistive, Level 2 Supervised, Level 3 Autonomous
   • Governance Gate (Mandatory): Action risk classification triggering mandatory human confirmation
3. Configure optional operational parameters and localized tags as required:
   • Context Truncation Limit (Optional): Maximum LLM token window before historical pruning
   • Fallback Model Route (Optional): Secondary LLM provider if primary endpoint is unresponsive
   • Webhook Authentication Token (Optional): Secure shared secret for n8n workflow callback
4. Click the Save icon in the control bar to commit the record.

{screenshot:prompt-governance-and-system-instructions-record-creation-form}

Step 3: Validating and Processing Transactions#

• State Transitions: Execute operational state progression (Prompt Ingestion → Reasoning Chain → Action Proposal → Human Gate Approved → Executed).
• Ledger & System Impact: Financial and inventory transactions executed by autonomous agents generate strict audit log entries tagging the AI Agent ID and human approver credentials.
• Audit Verification: Automated recording of author ID, timestamp, and immutable revision history.

{screenshot:prompt-governance-and-system-instructions-transaction-processing-state}

Prompt Governance Framework#

Enterprise AI reliability depends upon rigorous prompt engineering and strict system instructions. SmartERP manages system prompts as version-controlled database records (ashewa.ai.prompt.template).

Standard System Prompt Anatomy#

Every SmartERP Copilot system instruction includes four mandatory architectural components:

1. Role & Identity Definition: Clear mandate specifying the copilot's domain focus and professional tone.
2. Operational Scope & Boundaries: Explicit enumeration of permitted and prohibited activities (e.g., 'Never commit financial journal entries directly').
3. Statutory Compliance Directives: Instructions to enforce Ethiopian statutory laws (tax rates, overtime multipliers, store voucher rules).
4. Output Formatting Constraints: Strict JSON schema requirements for tool calls and clean markdown tables for human consumption.

4. Advanced Configuration Points#

• Automated Actions & Scheduled Tasks (Crons):
  • AI Bridge: Heartbeat check for n8n orchestrator nodes (ashewa_ai.ir_cron_heartbeat): Runs every 10 minutes to verify gateway connectivity
  • AI Governance: Auto-archive cleared approval queue tickets (ashewa_ai.ir_cron_archive_audit): Runs daily at midnight to maintain governance ledger performance
• Custom Print Templates & Reports:
  • AI Agent Operational Execution Audit Trail
  • LLM Token Consumption & Cost Allocation Schedule
  • Human-in-the-Loop Decision & Override Ledger
  • Agent Workflow Error & Latency Analysis
• Multi-Company Rules: Record-level security isolation under Company 1 (Ashewa Technology Solutions S.C.), preventing cross-company ledger leakage across parent and subsidiary legal entities.

5. Comprehensive Troubleshooting Matrix#