Your incident is already
diagnosed before you ask.
ContextOps is a Slack-native AI agent that reads your live AWS logs, remembers past fixes, diagnoses the outage, and writes the postmortem, turning a 2 AM scramble into a ten second conversation.
This is a Slack agent, not a website. It runs inside Slack. Judges: the sandbox workspace has been shared with your review addresses. Add the ContextOps app, mention it in a channel or open its DM, and describe an incident.
What is ContextOps?
ContextOps is an AI assistant that works inside Slack to help engineers fix production problems quickly. Instead of opening 10 different websites and dashboards, engineers can simply chat with the bot in Slack, and it does most of the investigation for them.
Think of it like ChatGPT + AWS monitoring + company memory + automation, all inside Slack.
The Problem It Solves
Normally, when a website or application breaks:
- Engineers open CloudWatch to check logs.
- Search old Slack messages.
- Read runbooks (repair guides).
- Ask teammates.
- Try to remember how they fixed it last time.
This wastes valuable time. ContextOps brings everything into one Slack conversation, saving minutes during an outage.
Example
Imagine your payment service suddenly starts giving 500 Internal Server Errors. Instead of doing all the investigation yourself, you type:
@ContextOps payment API returning 500 errorsThe bot replies with something like:
🔴 Severity: High
Possible cause: Database connection exhausted
Impact: Customers cannot make payments
Similar incident: Happened 3 weeks ago
Recommended fix: Restart database connection pool
Confidence: 93%Features Explained
1. Conversational Investigation
You simply chat with the bot. Type "Payment API is failing" and it responds with how serious it is, what probably broke, root cause, and what to do next. No need to search manually.
2. Live AWS Diagnostics
The bot automatically checks AWS CloudWatch Logs. It fetches errors, timeouts, and HTTP 500s from the last 30 minutes, removing duplicates so they're easier to read.
3. Slack Knowledge Search
The bot searches your company's Slack history. It finds only relevant past incidents and provides links to the original conversations.
4. AI Reasoning
Using Amazon Bedrock (Nova Pro), it combines current logs, old incidents, and Slack conversations to answer: "What's most likely broken?"
5. Learning Loop
Every incident gets saved. If a fix works today, ContextOps remembers it next month and suggests the exact same verified fix automatically.
6. Proactive Detection
ContextOps listens for AWS CloudWatch alarms. If CPU is too high or a service is down, it creates the incident in Slack before anyone asks.
7. Interactive Buttons
Slack shows buttons like Generate Report, Run Runbook, Acknowledge, Resolve, and Escalate. Clicking them performs the action instantly.
8. Automated Runbooks
Press "Run Runbook" and the bot performs recovery steps (like restarting Redis or verifying connections) and reports the result.
9. Incident Reports
After a fix, ContextOps automatically creates a professional report and a Slack Canvas postmortem so the team has a permanent record.
10. Mission Control Dashboard
The Slack Home tab shows open incidents, resolved incidents, MTTR, and severity breakdowns with buttons to scan logs on demand.
System Architecture
Complete Workflow
Tech Stack Explained
| Technology | Purpose |
|---|---|
| Python | Main programming language |
| FastAPI | Backend API server |
| Slack Bolt | Connects the app with Slack |
| Slack Search API | Searches old Slack messages |
| Amazon Bedrock (Nova Pro) | AI reasoning |
| AWS CloudWatch Logs | Reads application logs |
| Amazon S3 | Stores reports/files |
| AWS IAM | Manages AWS permissions securely |
| SQLite | Stores incidents and history |
| EC2 | Runs the application |
| Nginx | Web server and reverse proxy |
| systemd | Keeps the app running automatically |