Olympus Chat: Unified Collaboration Platform

Olympus Chat is a product-grade, native Slack and PagerDuty replacement built into Olympus Cloud. It merges Staff Communication and Custom Alerting into a single cohesive experience.

Overview

Attribute	Value
Status	GA
Primary Users	All staff, operations, support
Replaces	Slack, PagerDuty, external chat tools
Latency Target	Sub-100ms message delivery

┌─────────────────────────────────────────────────────────────────┐
│  OLYMPUS CHAT ARCHITECTURE                                       │
├─────────────────────────────────────────────────────────────────┤
│                                                                  │
│  ┌──────────────┐   ┌──────────────┐   ┌──────────────┐        │
│  │ Olympus Chat │   │ Olympus Alert│   │ Olympus      │        │
│  │ Core         │   │ Incidents    │   │ Connect      │        │
│  │              │   │              │   │ Integrations │        │
│  │ • Channels   │   │ • On-Call    │   │ • Slack      │        │
│  │ • Threads    │   │ • Escalation │   │ • Teams      │        │
│  │ • DMs        │   │ • Runbooks   │   │ • Webhooks   │        │
│  └──────┬───────┘   └──────┬───────┘   └──────┬───────┘        │
│         │                  │                  │                 │
│         └────────────┬─────┴──────────────────┘                │
│                      │                                          │
│              ┌───────┴───────┐                                 │
│              │   Maximus AI  │                                 │
│              │  Operations   │                                 │
│              │     Bot       │                                 │
│              └───────────────┘                                 │
│                                                                  │
└─────────────────────────────────────────────────────────────────┘

---

Core Pillars

1. Olympus Chat Core (Collaboration)

Real-time messaging with enterprise features.

Feature	Description
Workspace Hierarchy	Tenant → Brand → Location → Department
Channels	Public and private channels with threading
Direct Messages	One-on-one encrypted messaging
Threading	Organized conversations within channels
Read Receipts	Real-time delivery and read status
Presence	Online/offline/away status indicators
Rich Attachments	Images, files, code snippets
Global Search	Full-text search across messages

Related: Staff Chat Guide

2. Olympus Alert (Incident Management)

PagerDuty replacement with integrated incident response.

Feature	Description
Dynamic Incident Channels	Auto-created per alert
On-Call Schedules	Visual schedule management
Escalation Policies	Configurable escalation paths
In-Chat Controls	Acknowledge, Resolve, Escalate buttons
Runbook Integration	Auto-attach relevant runbooks
Post-Mortems	Built-in incident review workflows

Related: Incident Response Runbook

3. Olympus Connect (Integrations)

External bridges and webhook ingestion.

Integration	Direction	Description
Slack Bridge	Bidirectional	Mirror messages to/from Slack
Teams Bridge	Bidirectional	Microsoft Teams integration
Discord Bridge	Outbound	Send to Discord channels
Webhook Ingestion	Inbound	Accept alerts from any monitoring tool
Twilio	Outbound	SMS/voice notifications
SendGrid	Outbound	Email notifications
Push	Outbound	Mobile push notifications

4. Maximus AI Bot (Intelligence)

AIOps-driven operational intelligence.

Capability	Description
Natural Language Queries	"What's the status of order processing?"
Incident Summarization	Auto-generate incident summaries
Suggested Actions	AI-recommended remediation steps
Real-time Translation	21 language support
Voice-to-Text	Transcribe voice messages
L1/L2 Automation	Auto-resolve known issues

Related: Dev Agent Guide

---

Workspace Hierarchy

Olympus Chat follows the Olympus Cloud multi-tenant hierarchy:

Tenant (NebusAI)
├── Brand (Restaurant Revolution)
│   ├── Location (Downtown NYC)
│   │   ├── Department (Kitchen)
│   │   │   ├── #kitchen-general
│   │   │   └── #kitchen-prep
│   │   └── Department (Front-of-House)
│   │       ├── #foh-general
│   │       └── #foh-reservations
│   └── Location (Midtown NYC)
│       └── ...
└── Brand (Olympus Cloud)
    └── ...

Channel Types

Type	Visibility	Use Case
Public	Anyone in scope	General announcements
Private	Invited members only	Sensitive discussions
Incident	Auto-created	Per-incident communication
Support	Customer + Support	Customer support threads
DM	Two people	Private conversations

---

On-Call Management

Schedule Configuration

schedule:
  name: "Backend On-Call"
  timezone: "America/New_York"
  rotations:
    - type: weekly
      start: "2026-01-01T09:00:00"
      members:
        - user: "alice@nebusai.com"
          order: 1
        - user: "bob@nebusai.com"
          order: 2
  overrides:
    - user: "charlie@nebusai.com"
      start: "2026-01-15T00:00:00"
      end: "2026-01-16T00:00:00"
      reason: "Covering for Alice"

Escalation Policies

Level	Target	Wait Time	Action
L1	Primary on-call	0 min	Page via push + SMS
L2	Secondary on-call	5 min	Page if L1 doesn't ack
L3	Team lead	15 min	Page if L2 doesn't ack
L4	Engineering manager	30 min	Critical escalation

---

Incident Workflow

Auto-Created Incident Channels

When an alert fires, Olympus Chat automatically:

1. Creates dedicated channel: #incident-2026-01-24-database-high-cpu
2. Invites on-call responders: Based on escalation policy
3. Attaches runbook: If matched by alert metadata
4. Posts alert summary: Including metrics and context
5. Starts incident timeline: Auto-logs all actions

In-Chat Incident Controls

┌────────────────────────────────────────────────────────────┐
│ 🔴 INCIDENT: Database CPU > 90% (Critical)                │
│                                                            │
│ Started: 14:32 UTC | Duration: 23m                        │
│ Assigned: alice@nebusai.com                               │
│                                                            │
│ [Acknowledge] [Resolve] [Escalate] [Snooze 30m]           │
│                                                            │
│ 📋 Runbook: database-high-cpu.md                          │
│ 📊 Metrics: cloud.spanner.cpu_utilization                 │
└────────────────────────────────────────────────────────────┘

---

Integration with ACP

Olympus Chat integrates with the ACP AI Router for intelligent message processing:

Model Tier	Use Case	Cost
T1 (Llama 4 Scout)	Simple greetings, status checks	FREE
T2 (Gemini 2.0 Flash)	Message translation, search	$0.10/M
T3 (Gemini 3 Flash)	Incident summarization	$0.50/M
T4 (Claude Haiku 4.5)	Complex queries	$1.00/M
T5 (Claude Sonnet 4.5)	Strategic recommendations	$3.00/M

---

API Reference

WebSocket Connection

// Connect to Olympus Chat WebSocket
const ws = new WebSocket('wss://api.olympuscloud.ai/ws/chat');

ws.onopen = () => {
  ws.send(JSON.stringify({
    type: 'auth',
    token: 'Bearer <jwt_token>'
  }));
};

ws.onmessage = (event) => {
  const message = JSON.parse(event.data);
  switch (message.type) {
    case 'message':
      handleNewMessage(message);
      break;
    case 'presence':
      handlePresenceUpdate(message);
      break;
    case 'incident':
      handleIncidentUpdate(message);
      break;
  }
};

Send Message

POST /api/v1/chat/channels/{channel_id}/messages
Authorization: Bearer <token>
Content-Type: application/json

{
  "content": "Hello team!",
  "thread_id": "optional-thread-id",
  "attachments": []
}

Create Incident

POST /api/v1/incidents
Authorization: Bearer <token>
Content-Type: application/json

{
  "title": "Database High CPU",
  "severity": "critical",
  "service": "platform-service",
  "description": "CPU utilization exceeded 90%",
  "runbook_url": "/runbooks/database-high-cpu"
}

---

Configuration

Environment Variables

Variable	Description	Default
OLYMPUS_CHAT_WEBSOCKET_URL	WebSocket endpoint	wss://api.olympuscloud.ai/ws/chat
OLYMPUS_CHAT_MAX_MESSAGE_SIZE	Max message bytes	65536
OLYMPUS_CHAT_MESSAGE_RETENTION_DAYS	Message retention	365
OLYMPUS_CHAT_INCIDENT_AUTO_RESOLVE_HOURS	Auto-resolve timeout	24

Feature Flags

Flag	Description	Default
olympus_chat.enabled	Enable Olympus Chat	true
olympus_chat.voice_transcription	Voice-to-text	true
olympus_chat.ai_translation	Auto-translation	true
olympus_chat.external_bridges	Slack/Teams bridges	false

---

Performance Targets

Metric	Target	Actual
Message Latency (p50)	<50ms	35ms
Message Latency (p99)	<100ms	85ms
Incident Creation	<500ms	320ms
Search Response	<200ms	150ms
Concurrent Users	10,000+	Tested

---

Security

Encryption

info

End-to-end encryption (E2E) is available for sensitive channels but must be explicitly enabled. When E2E is active, server-side search and AI features are unavailable for those channels since the server cannot read the content.

• In Transit: TLS 1.3 for all connections
• At Rest: AES-256 encryption for messages
• E2E Option: Available for sensitive channels

Compliance

• Message Retention: Configurable per tenant
• Audit Logging: All actions logged
• GDPR: Data export and deletion support
• SOC 2: Type II compliant

---

Migration from External Tools

From Slack

1. Export Slack workspace data
2. Use Olympus Connect import wizard
3. Map channels to Olympus Chat hierarchy
4. Configure bidirectional bridge (optional)
5. Gradual migration with bridge active

From PagerDuty

warning

Run Olympus Alert in parallel with PagerDuty for at least two on-call rotations before cutting over. This ensures escalation policies, schedule overrides, and webhook integrations are validated under real incident conditions.

1. Export schedules and escalation policies
2. Import into Olympus Alert
3. Configure webhook ingestion
4. Update monitoring tools to send to Olympus Chat
5. Run parallel for verification period

---

Related Documentation

• Staff Chat Guide - End-user messaging guide
• Alerting Configuration - Alert setup
• Incident Response Runbook - Incident procedures
• ACP Router - AI model routing
• Dev Agent - Engineering AI assistant

---

Olympus Chat: Unified Collaboration & Incident Management Platform Last Updated: 2026-02-19