OpsLens

Autonomous Incident Response Orchestrator powered by Notion MCP

OpsLens transforms Notion into an AI-powered incident command center. It ingests alerts from monitoring tools, runs a pipeline of specialized AI agents for triage, correlation, remediation, and postmortem generation, and writes every finding back to Notion as structured, searchable knowledge. Engineers interact through a real-time dashboard or directly in Notion. The system watches for human edits and reacts, creating a true human-in-the-loop incident response workflow.

Built for the Notion MCP Challenge on DEV.to.

Table of Contents

• Problem Statement
• How It Works
• Architecture
• Use Cases
• Features
• Tech Stack
• Project Structure
• Getting Started
• Configuration
• API Reference
• Webhook Integration
• Slash Commands
• Docker Deployment
• Development
• Author
• License

Problem Statement

When a production incident fires at 3 AM, the on-call engineer faces a wall of context switching: triage the alert, search for past incidents, find the runbook, notify stakeholders, check recent deployments, and document everything for the postmortem. Each step is manual, scattered across different tools, and prone to human error under pressure.

OpsLens eliminates this friction. It receives the alert, runs five specialized AI agents in sequence, writes every finding to Notion, and presents the engineer with a clear picture: what happened, what is affected, what to do next, and who to call. The engineer stays in control. The AI handles the grunt work.

How It Works

Alert Source                    OpsLens Backend                         Notion (via MCP)
-----------                    ---------------                         ----------------

Prometheus  ----+
Grafana     ----+---> Webhook --> Normalize --> Dedup/Group --> Create Incident Page
PagerDuty   ----+         |                                         |
Slack       ----+         |                                         |
Manual      ----+         v                                         v
                    Agent Pipeline                            Page Comments
                    1. Triage Agent ------> severity, category ------> comment
                    2. Correlation Agent -> past incidents, patterns -> comment
                    3. Remediation Agent -> runbook steps, fixes -----> comment
                    4. Comms Agent -------> escalation, notifications
                    5. Postmortem Agent --> blameless postmortem page
                          |
                          v
                    Real-time Dashboard <------- WebSocket <------- State Changes
                          |
                          v
                    Notion Watcher (polls) --> detect human edits --> re-trigger agents

1. An alert arrives via webhook (Prometheus, Grafana, PagerDuty, or custom source).
2. The normalizer converts it to a canonical format. The deduplicator checks for duplicates. The alert grouper checks if it belongs to an existing incident.
3. A Notion page is created in the Incidents database with full alert metadata.
4. Five AI agents run in sequence. Each agent uses Notion MCP tools to search the workspace, fetch runbooks, find past incidents, and cross-reference connected tools (Slack, Google Drive,

... View full README on GitHub