Adoption Path: Day-1 to Day-30

A structured 30-day implementation guide to take you from first action to audit-ready production deployment.

Timeline Overview

DAY 1          DAY 7           DAY 14          DAY 21          DAY 30
│              │               │               │               │
▼              ▼               ▼               ▼               ▼
┌────────┐   ┌────────┐    ┌────────┐     ┌────────┐     ┌────────┐
│ FIRST  │   │ FIRST  │    │ TEAM   │     │ PROD   │     │ AUDIT  │
│ ACTION │   │ POLICY │    │ ROLLOUT│     │ DEPLOY │     │ READY  │
└────────┘   └────────┘    └────────┘     └────────┘     └────────┘
   │              │               │               │               │
   ▼              ▼               ▼               ▼               ▼
  SDK          Custom         5+ Agents      Production      Compliance
 Setup         Rules          Governed       Traffic         Reports

Success Metrics by Phase

Metric	Day 1	Day 7	Day 14	Day 21	Day 30
Agents Registered	1	1	5+	10+	All production
Policies Active	0	2	5+	10+	Full coverage
Actions Evaluated	10	100	500+	2000+	All agent actions
Team Members Trained	1	1	5+	All operators	All stakeholders
Approval Workflows	0	1	3+	5+	Risk-appropriate
SIEM Integration	No	No	Started	Connected	Streaming
Compliance Reports	No	No	No	Draft	Audit-ready

Week 1: Foundation

Day 1 — First Governed Action

Goal: Submit your first action to ASCEND and see governance in action.

Time Required: 30 minutes

What You'll Do:

Get API credentials

# Log in to the dashboard
open https://pilot.owkai.app

# Navigate to Settings → API Keys → Generate New Key
# Copy the key (shown only once)

# Set environment variable
export ASCEND_API_KEY="owkai_admin_xxxxxxxxxxxx"

Install the SDK
```
pip install requests python-dotenv
```

Submit your first action

import os
import requests

response = requests.post(
    "https://pilot.owkai.app/api/v1/actions/submit",
    headers={
        "Authorization": f"Bearer {os.getenv('ASCEND_API_KEY')}",
        "Content-Type": "application/json"
    },
    json={
        "agent_id": "test-agent-001",
        "agent_name": "My First Agent",
        "action_type": "query",
        "description": "Test action - read public data",
        "tool_name": "test_tool"
    }
)
print(response.json())

View the result in the dashboard
- Navigate to Activity Feed
- Find your action
- Review the risk score

What You'll Learn:

How action evaluation works
Risk scoring basics (0-100 scale)
Dashboard navigation

Success Criteria:

API key generated
First action submitted
Action visible in dashboard
Risk score displayed

Day 2-3 — SDK Integration

Goal: Integrate ASCEND SDK into a real AI agent.

Time Required: 2-4 hours

What You'll Do:

Choose your integration pattern

For new agents (recommended):

from ascend_client import ASCENDClient

client = ASCENDClient()

def my_agent_action(action_type, resource, description):
    # Governance check before action
    result = client.submit_action(
        agent_id="my-production-agent",
        agent_name="Production Agent",
        action_type=action_type,
        description=description,
        tool_name="my_tool",
        resource=resource
    )

    if result['status'] == 'approved':
        return execute_action()
    elif result['status'] == 'pending_approval':
        return wait_for_approval(result['id'])
    else:
        return handle_denial(result)

Test with different risk levels

Action Type	Expected Risk	Expected Decision
`query`	Low (10-25)	Auto-approve
`data_access`	Medium (30-50)	Auto-approve
`database_write`	Medium-High (45-65)	May require approval
`credential_access`	High (70-85)	Requires approval

Handle pending approvals

if result['status'] == 'pending_approval':
    print(f"Action {result['id']} requires approval")

    # Poll for decision
    while True:
        status = client.get_action_status(result['id'])
        if status['status'] != 'pending_approval':
            break
        time.sleep(5)

What You'll Learn:

SDK integration patterns
Handling different decision types
Approval workflow basics

Success Criteria:

SDK integrated into agent
Low-risk actions auto-approved
High-risk actions require approval
Pending approvals handled correctly

Day 4-5 — Dashboard Exploration

Goal: Master the ASCEND dashboard for monitoring and operations.

Time Required: 1-2 hours

What You'll Do:

Explore key sections

Section	Purpose	Key Actions
Activity Feed	Real-time action stream	Filter, search, review
Agent Registry	Registered agents	View status, suspend
Alerts	Security notifications	Acknowledge, investigate
Policies	Governance rules	View, create (Day 7)
Analytics	Metrics and trends	Review risk distribution

Set up your first alert
- Navigate to Alerts → Alert Rules
- Create: "High Risk Action" (risk > 70)
- Enable email notification
Review the Analytics dashboard
- Action volume by day
- Risk distribution
- Approval rates
- Top agents by activity

What You'll Learn:

Dashboard navigation
Alert configuration
Analytics interpretation

Success Criteria:

Visited all main sections
Alert rule created
Analytics data reviewed

Day 6-7 — First Custom Policy

Goal: Create your first governance policy.

Time Required: 2-3 hours

What You'll Do:

Design your policy

Example: Block database deletes in production

IF action_type = "database_delete"
AND environment = "production"
THEN require_approval from ["dba-team"]

Create via API

curl -X POST https://pilot.owkai.app/api/unified-governance/policies \
  -H "Authorization: Bearer $ASCEND_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Production Database Delete Protection",
    "description": "Require DBA approval for production deletes",
    "is_active": true,
    "priority": 80,
    "conditions": {
      "action_type": "database_delete",
      "environment": "production"
    },
    "effect": "require_approval",
    "approvers": ["dba-team"]
  }'

Test the policy

result = client.submit_action(
    agent_id="test-agent",
    action_type="database_delete",
    resource="production_users_table",
    description="Delete inactive users",
    context={"environment": "production"}
)
# Should return: pending_approval

Verify in dashboard
- Navigate to Policies
- Find your new policy
- Check policy evaluation logs

What You'll Learn:

Policy condition syntax
Priority and conflict resolution
Policy testing workflow

Success Criteria:

Policy created successfully
Policy triggers on matching actions
Approval required as expected
Policy visible in dashboard

Week 2: Expansion

Day 8-10 — Policy Design Workshop

Goal: Design comprehensive policies for your organization.

Time Required: 4-6 hours

What You'll Do:

Audit your agent actions
- Review Activity Feed for past 7 days
- Identify action types in use
- Note high-risk patterns

Create policy categories

Category	Example Policy	Priority
Data Protection	Require approval for PII access	90
Financial Controls	Block large transactions without approval	85
Infrastructure	Require approval for production changes	80
Compliance	Log all HIPAA-relevant actions	75

Build your policy set

Start with these essential policies:
- Block credential_access in production
- Require approval for database_delete
- Alert on risk score > 80
- Auto-approve query actions with risk < 30

What You'll Learn:

Policy design patterns
Balancing security and productivity
Conflict resolution

Success Criteria:

5+ policies created
Policies tested and active
No conflicting rules

Day 11-12 — Approval Workflows

Goal: Configure multi-level approval workflows.

Time Required: 3-4 hours

What You'll Do:

Design workflow tiers

Risk 0-30:   Auto-approve
Risk 31-60:  Team lead approval
Risk 61-80:  Manager + Security approval
Risk 81+:    VP approval required

Create workflow configuration

{
  "name": "Tiered Approval Workflow",
  "trigger_conditions": {
    "min_risk": 61,
    "max_risk": 80
  },
  "steps": [
    {
      "name": "Manager Review",
      "approvers": ["manager-group"],
      "required_approvals": 1,
      "timeout_hours": 4
    },
    {
      "name": "Security Review",
      "approvers": ["security-team"],
      "required_approvals": 1,
      "timeout_hours": 8
    }
  ],
  "escalation": {
    "enabled": true,
    "timeout_hours": 24,
    "escalation_approvers": ["vp-operations"]
  }
}

Set up notifications
- Slack webhook for pending approvals
- Email for escalations
- PagerDuty for critical alerts

What You'll Learn:

Multi-stage approvals
Timeout and escalation
Notification integration

Success Criteria:

Workflow created and tested
Notifications configured
Escalation works correctly

Day 13-14 — Team Onboarding

Goal: Train your team on ASCEND operations.

Time Required: 4-6 hours

What You'll Do:

Create user accounts

Navigate to Admin → Users
Invite team members
Assign appropriate roles

Role	Permissions	Typical Users
Viewer	Read-only dashboard	Stakeholders
Analyst	View + alerts	Security analysts
Operator	Approve/deny actions	Team leads
Manager	Configure policies	Security managers
Admin	Full access	Platform admins

Conduct training sessions

Session 1: Dashboard Overview (30 min)
- Activity Feed navigation
- Alert management
- Basic reporting
Session 2: Approval Workflows (30 min)
- How to review pending actions
- Approve/deny process
- Escalation handling
Session 3: Policy Management (45 min)
- Policy creation
- Testing and validation
- Conflict resolution
Document procedures
- Approval SLAs
- Escalation contacts
- Emergency procedures

What You'll Learn:

RBAC configuration
Team training best practices
Operational procedures

Success Criteria:

5+ team members onboarded
Training sessions completed
Procedures documented

Week 3: Scale

Day 15-17 — Multi-Agent Deployment

Goal: Expand governance to all AI agents.

Time Required: 6-8 hours

What You'll Do:

Inventory all agents

Agent Type Risk Level Priority
Customer Support Bot Query-only Low P3
Financial Advisor Transactions High P1
Data Pipeline Read/Write Medium P2
Code Assistant Code execution High P1
Prioritize integration
- P1: High-risk agents (Day 15-16)
- P2: Medium-risk agents (Day 17)
- P3: Low-risk agents (Week 4)

Agent	Type	Risk Level	Priority
Customer Support Bot	Query-only	Low	P3
Financial Advisor	Transactions	High	P1
Data Pipeline	Read/Write	Medium	P2
Code Assistant	Code execution	High	P1

Register each agent

# For each agent
client = ASCENDClient(
    api_key=os.getenv("ASCEND_API_KEY"),
    agent_id="financial-advisor-001",
    agent_name="Financial Advisor Bot"
)

Configure agent-specific policies
- Each agent may need custom risk thresholds
- Define allowed action types per agent
- Set notification preferences

What You'll Learn:

Multi-agent management
Agent-specific configuration
Risk-based prioritization

Success Criteria:

5+ agents registered
Agent-specific policies active
All P1 agents governed

Day 18-19 — Alert Configuration

Goal: Configure comprehensive alerting.

Time Required: 3-4 hours

What You'll Do:

Define alert categories

Alert Type	Trigger	Notification	Response
Critical Risk	Risk > 90	PagerDuty	Immediate
High Risk	Risk > 70	Slack + Email	15 min
Policy Violation	Blocked action	Slack	Review
Anomaly Detected	Unusual pattern	Email	Investigate

Configure SIEM integration

curl -X POST https://pilot.owkai.app/api/siem-integration/configure \
  -H "Authorization: Bearer $ASCEND_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "siem_type": "splunk",
    "host": "your-splunk-instance.com",
    "port": 8088,
    "api_token": "your-hec-token",
    "index": "ascend_events"
  }'

Test alert delivery
- Submit a high-risk test action
- Verify alert received in all channels
- Check SIEM event ingestion

What You'll Learn:

Alert rule configuration
SIEM integration
Incident response workflows

Success Criteria:

Alert rules configured
Notifications tested
SIEM receiving events

Day 20-21 — Production Planning

Goal: Plan production deployment.

Time Required: 4-6 hours

What You'll Do:

Production readiness checklist
- All P1/P2 agents integrated
- Policies reviewed and approved
- Approval workflows tested
- Alerting configured
- Team trained
- Runbooks documented
- Rollback plan defined

Define production configuration

# Production SDK configuration
client = ASCENDClient(
    api_key=os.getenv("ASCEND_PROD_API_KEY"),
    fail_mode=FailMode.CLOSED,  # Block if ASCEND unavailable
    timeout=30,
    retries=3
)

Create deployment plan

Phase Actions Rollback Trigger
Phase 1 10% traffic Error rate > 1%
Phase 2 50% traffic Error rate > 0.5%
Phase 3 100% traffic Error rate > 0.1%

Phase	Actions	Rollback Trigger
Phase 1	10% traffic	Error rate > 1%
Phase 2	50% traffic	Error rate > 0.5%
Phase 3	100% traffic	Error rate > 0.1%

What You'll Learn:

Production configuration
Deployment planning
Rollback procedures

Success Criteria:

Checklist complete
Configuration reviewed
Deployment plan approved

Week 4: Production

Day 22-24 — Production Deployment

Goal: Deploy ASCEND to production traffic.

Time Required: 6-8 hours

What You'll Do:

Phase 1: Canary deployment (10%)
- Deploy to subset of agents/traffic
- Monitor error rates
- Verify latency acceptable
Phase 2: Expanded deployment (50%)
- Increase traffic percentage
- Monitor approval workflow performance
- Check alert volume
Phase 3: Full deployment (100%)
- Route all traffic through ASCEND
- Enable all policies
- Activate all alerts
Monitor key metrics

Metric Target Alert Threshold
Error rate < 0.1% > 0.5%
p99 latency < 150ms > 300ms
Approval SLA < 4hr > 8hr

Metric	Target	Alert Threshold
Error rate	< 0.1%	> 0.5%
p99 latency	< 150ms	> 300ms
Approval SLA	< 4hr	> 8hr

What You'll Learn:

Production deployment
Canary releases
Production monitoring

Success Criteria:

100% traffic through ASCEND
Error rate < 0.1%
No production incidents

Day 25-27 — Monitoring Setup

Goal: Establish comprehensive production monitoring.

Time Required: 4-6 hours

What You'll Do:

Set up dashboards

Executive Dashboard
- Total actions governed (24h)
- Risk distribution
- Approval rate
- Top agents by volume
Operations Dashboard
- Real-time action stream
- Pending approvals count
- Alert status
- Error rate

Configure Datadog/CloudWatch integration

# Export metrics to Datadog
DiagnosticAuditLog.to_datadog_metrics() → [
  {
    "metric": "ascend.actions.total",
    "type": "count",
    "points": [[timestamp, count]],
    "tags": ["org_id:123", "environment:production"]
  }
]

Create runbooks
- High error rate response
- Approval backlog handling
- Agent suspension procedure
- Incident escalation

What You'll Learn:

Production monitoring
Dashboard creation
Operational runbooks

Success Criteria:

Dashboards created
Metrics flowing
Runbooks documented

Day 28-30 — Compliance Audit

Goal: Generate audit-ready compliance documentation.

Time Required: 6-8 hours

What You'll Do:

Generate compliance reports

curl -X POST https://pilot.owkai.app/api/compliance-export/exports \
  -H "Authorization: Bearer $ASCEND_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "framework": "soc2",
    "report_type": "audit_log",
    "export_format": "json",
    "start_date": "2026-01-01",
    "end_date": "2026-01-31"
  }'

Review audit trail
- Verify all actions logged
- Check policy enforcement logs
- Validate approval trails
Document compliance posture

Framework Control Status Evidence
SOC 2 CC6.1 Logical access Complete Audit logs
SOC 2 CC6.7 Information disposal Complete Retention policies
HIPAA 164.312(b) Audit controls Complete Activity monitoring
Prepare audit package
- Policy documentation
- Approval workflow evidence
- Risk assessment records
- Training completion records

Framework	Control	Status	Evidence
SOC 2 CC6.1	Logical access	Complete	Audit logs
SOC 2 CC6.7	Information disposal	Complete	Retention policies
HIPAA 164.312(b)	Audit controls	Complete	Activity monitoring

What You'll Learn:

Compliance reporting
Audit preparation
Evidence collection

Success Criteria:

Compliance reports generated
Audit trail verified
Documentation complete
Audit-ready package prepared

Quick Reference

Day-by-Day Checklist

Day	Milestone	Time	Key Deliverable
1	First Action	30 min	Action in dashboard
2-3	SDK Integration	2-4 hr	Agent connected
4-5	Dashboard	1-2 hr	Alerts configured
6-7	First Policy	2-3 hr	Policy active
8-10	Policy Design	4-6 hr	5+ policies
11-12	Workflows	3-4 hr	Approvals working
13-14	Team Onboarding	4-6 hr	Team trained
15-17	Multi-Agent	6-8 hr	5+ agents
18-19	Alerts	3-4 hr	SIEM connected
20-21	Planning	4-6 hr	Plan approved
22-24	Production	6-8 hr	100% traffic
25-27	Monitoring	4-6 hr	Dashboards live
28-30	Compliance	6-8 hr	Audit-ready

Support Resources

Enterprise Support: support@ascendowkai.com
Documentation: https://docs.ascendowkai.com
Status Page: https://pilot.owkai.app/health
Slack Community: Contact your account manager

Next Steps

After Day 30, continue with:

Continuous Improvement
- Weekly policy reviews
- Monthly risk assessment
- Quarterly compliance audits
Advanced Features
- Smart Rules (ML-generated policies)
- A/B testing for policies
- Custom risk models
Scale
- Additional agents
- Cross-team rollout
- Enterprise SSO integration

Last Updated: 2026-01-21

Timeline Overview​

Success Metrics by Phase​

Week 1: Foundation​

Day 1 — First Governed Action​

Day 2-3 — SDK Integration​

Day 4-5 — Dashboard Exploration​

Day 6-7 — First Custom Policy​

Week 2: Expansion​

Day 8-10 — Policy Design Workshop​

Day 11-12 — Approval Workflows​

Day 13-14 — Team Onboarding​

Week 3: Scale​

Day 15-17 — Multi-Agent Deployment​

Day 18-19 — Alert Configuration​

Day 20-21 — Production Planning​

Week 4: Production​

Day 22-24 — Production Deployment​

Day 25-27 — Monitoring Setup​

Day 28-30 — Compliance Audit​

Quick Reference​

Day-by-Day Checklist​

Support Resources​

Next Steps​

Timeline Overview

Success Metrics by Phase

Week 1: Foundation

Day 1 — First Governed Action

Day 2-3 — SDK Integration

Day 4-5 — Dashboard Exploration

Day 6-7 — First Custom Policy

Week 2: Expansion

Day 8-10 — Policy Design Workshop

Day 11-12 — Approval Workflows

Day 13-14 — Team Onboarding

Week 3: Scale

Day 15-17 — Multi-Agent Deployment

Day 18-19 — Alert Configuration

Day 20-21 — Production Planning

Week 4: Production

Day 22-24 — Production Deployment

Day 25-27 — Monitoring Setup

Day 28-30 — Compliance Audit

Quick Reference

Day-by-Day Checklist

Support Resources

Next Steps