71Governance health
AI Control Plane / Integrate

SDK & Integration

Integrate Aegis with your existing AI infrastructure. Connect model registries, observability tools, and data platforms to automate governance workflows.

Total Systems
0
connected
Active Now
0
systems online
Events Total
0
governance checks
Avg Latency
—
last 24h
QUICKSTART
# Install
pip install aegis-ai

# Initialize
from aegis import AegisClient

client = AegisClient(api_key="aeg_prod_xxxxxxxxxxxx")

# Log a governance decision
response = client.governance.check(
    agent_id="fraud-detection-v2",
    action="transaction_review",
    context={
        "confidence_score": 0.94,
        "risk_score": 0.12,
        "transaction_amount": 4200.00
    }
)

print(response.decision)      # ALLOW
print(response.pattern_type)  # BASELINE
print(response.latency_ms)    # 38
RESPONSE FORMAT

Every governance check returns a structured JSON response. Decisions are either ALLOW or DENY.

✓ ALLOW
{
  "decision": "ALLOW",
  "policy_id": "pol_sr117_001",
  "reason": "Within approved parameters",
  "latency_ms": 23
}
✗ DENY
{
  "decision": "DENY",
  "policy_id": "pol_sr117_002",
  "reason": "Risk score exceeds threshold (0.72 > 0.70)",
  "latency_ms": 18,
  "pattern_triggered": false
}
API KEYS
NameKeyStatusLast UsedCreated
No API keys yet. Click "+ Generate Key" to create one.
CONNECTED SYSTEMS
No systems connected yet.
Make your first governance check to see it appear here.
INTEGRATION GUIDES
WEBHOOKSPILOT TIER
Coming soon
Event Types
policy.violationmodel.approval_requiredaudit.generated
CONNECT YOUR TOOLS
Datadog
Available
Stream governance events to Datadog as custom metrics and logs. Monitor policy violations alongside your existing observability stack.
Databricks
Available
Register models trained in Databricks directly into the Aegis registry via API. Trigger governance checks on every model version.
Azure ML
Coming soon
Connect Azure ML model deployments to Aegis for automatic SR 11-7 audit trail generation.
SageMaker
Coming soon
Register AWS SageMaker endpoints in Aegis registry and apply governance policies at deployment time.
AGENT GOVERNANCE
Wire Aegis into any AI agent
Add a single policy check before any agent action. Aegis evaluates it against your active policies and returns ALLOW, BLOCK, or ESCALATE in real-time.
const check = await fetch('https://app.withaegis.ai/api/agent/policy/check', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({
    agent_id: 'your-agent-id',
    session_id: sessionId,
    action_type: 'tool_use',
    action_payload: { tool: toolName, args: toolArgs },
    company_id: 'your-company-id'
  })
});
const { decision } = await check.json();
if (decision === 'BLOCK') throw new Error('Blocked by Aegis policy');
if (decision === 'ESCALATE') await requestHumanApproval();
For GSIB deployments: Aegis deploys inside your Azure or AWS tenant. API calls route through your private endpoint. No data leaves your environment.
Ask AegisADMIN