Olly
AI agent by Coralogix
The AI Observability Agent for Instant System Insights
Olly is an autonomous AI observability agent that enables engineering teams to investigate logs, metrics, and traces through a conversational interface. It identifies root causes and suggests remediation steps while maintaining full transparency by showing its underlying queries.
Agent specs
The full spec card (context, pricing per token, modalities and more) on AntSpace. Sign up free to see the full spec card.
- Languages
- English
Category: AI Agents
- Interface
- Conversational Natural Language
- Data Access
- Logs, Metrics, Traces
- Query Languages Generated
- DataPrime, PromQL
- Deployment
- Built-in to Coralogix Platform
- Autonomous AI observability agent for logs, metrics, and traces
- Conversational natural language interface for system queries
- Displays exact DataPrime and PromQL queries for transparency
- Continuously scans telemetry to surface anomalies and root causes
Overview
Olly is Coralogix's autonomous AI observability agent designed to help engineering teams monitor and investigate system data through a conversational interface. It connects directly to the Coralogix Telemetry Lake, allowing users to ask natural language questions about logs, metrics, and traces to identify root causes of incidents. The agent continuously scans telemetry to surface anomalies and suggests actionable next steps for remediation. Olly emphasizes transparency by displaying the exact DataPrime and PromQL queries used to generate its insights. It is built to democratize observability, enabling both engineers and non-technical team members to understand complex system behaviors.
Investigation and Remediation
- Conversational Interface: Users can interact with logs, metrics, and traces through a natural language conversational interface to ask questions about system behavior.
- Autonomous Investigation: The agent continuously scans telemetry data to surface anomalies and identify likely root causes without manual investigation.
- Actionable Insights: Olly suggests specific next steps to help engineering teams resolve incidents and keep systems running smoothly.
Trust and Collaboration
- Transparent Architecture: Olly displays all underlying DataPrime and PromQL queries so users can verify correctness and understand the process.
- Traceable Logic: The agent makes outputs traceable and justifiable by providing visibility into the logic behind every result.
- Unified Data Access: The agent connects directly to the Coralogix Telemetry Lake to access a unified foundation of logs, metrics, and traces.
Use cases
- Engineering teams ask natural language questions to identify the root cause of application slowness or payment flow failures.
- DevOps engineers use Olly to continuously scan telemetry for anomalies without writing manual DataPrime or PromQL queries.
- Non-technical team members interact with system data to understand service health and performance through plain English.
- Incident responders verify AI-generated insights by reviewing the transparent DataPrime and PromQL logic provided by the agent.
Specifications
- Interface
- Conversational Natural Language
- Data Access
- Logs, Metrics, Traces
- Query Languages Generated
- DataPrime, PromQL
- Deployment
- Built-in to Coralogix Platform
How it works
- Connect Telemetry Data: Users connect Olly to their existing Coralogix Telemetry Lake containing logs, metrics, and traces.
- Query in Plain English: Users ask natural language questions about system performance or specific incidents through the conversational interface.
- Analyze and Verify: Olly scans the data, identifies anomalies, and provides insights along with the specific queries used to find them.
- Act on Insights: The agent suggests actionable next steps for remediation to help resolve the incident quickly.
FAQ
What is Olly?
Olly is a standalone AI observability agent that connects to your Coralogix telemetry data—logs, metrics, and traces—and allows you to interact with it through a conversational interface.
Is Olly's logic transparent?
Olly defaults to transparency by displaying all the exact DataPrime and PromQL queries used to generate its insights, allowing users to verify results and improve their own practices.
Who can use Olly?
Olly is built to democratize observability, empowering all team members—not just engineers—to understand and act on complex system behavior.
Added by AntSpace from public sources (coralogix.com).