A powerful skill to query Prometheus monitoring metrics, alerts, and cluster health directly through an AI agent.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install prometheus-query
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install prometheus-query using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Prometheus Monitoring?
This Prometheus integration provides a streamlined way to interact with your monitoring infrastructure. By leveraging Openclaw Skills, developers and DevOps engineers can quickly retrieve system health data, analyze performance bottlenecks, and monitor Kubernetes clusters without leaving their terminal or agent environment. It bridges the gap between raw monitoring data and actionable insights within your AI-driven workflows.
The skill supports both remote URL access and Kubernetes-based port-forwarding, making it versatile for various deployment architectures. Whether you are checking the status of a specific pod or analyzing global Nginx request rates, this tool simplifies the process of data extraction from Prometheus instances.
Prometheus Monitoring Use Cases
- Checking the health status of Kubernetes nodes and pods in real-time.
- Retrieving active firing or pending alerts from Prometheus to respond to incidents.
- Monitoring Nginx request rates and connection statistics for web performance tuning.
- Executing custom PromQL queries for specific application-level metrics.
- Troubleshooting system performance bottlenecks during high-load scenarios.
How Prometheus Monitoring Works
- The user specifies a Prometheus URL or uses the default localhost endpoint via the CLI script.
- The script establishes a connection to the Prometheus API, either directly over the network or through a Kubernetes port-forward session.
- A PromQL query or alert request is dispatched to the server using the configured parameters.
- Prometheus processes the request and returns the corresponding metric data, time-series results, or alert states.
- The skill formats the raw JSON response into a human-readable, developer-friendly output for immediate analysis by the user or agent.
Prometheus Monitoring Setup
Ensure you have Python installed and the necessary scripts available in your directory. For remote access to a Prometheus instance, use the following command:
python scripts/query_prometheus.py --prom-url "https://prometheus.example.com" --query "up"
If your Prometheus instance is only accessible within a Kubernetes cluster, first establish a port-forward in a separate terminal:
kubectl port-forward -n prometheus svc/prometheus 9090:9090
Then, query using the local address:
python scripts/query_prometheus.py --prom-url "http://localhost:9090" --query "rate(http_requests_total[5m])"
Prometheus Monitoring Data Schema & Taxonomy
The skill returns structured data based on the specific Prometheus query type. Openclaw Skills ensure the output is parsed correctly for the agent's context:
| Data Component | Description |
|---|---|
| Instant Query | Returns current scalar values and metric labels for a point-in-time snapshot. |
| Range Query | Returns time-series data including timestamps and values for trend analysis. |
| Alert Data | Detailed lists including alert names, states (firing or pending), and metadata labels. |
| Configuration | Metadata such as evaluation time, resolution steps, and timeout settings. |
Prometheus Monitoring Advanced Features
- Support for both instant and range queries for deep historical data analysis.
- Integrated alert monitoring to track firing and pending states across the cluster.
- Flexible connection modes allowing for direct cloud URL or local Kubernetes port-forwarding.
- Full PromQL support for complex data aggregations, filtering, and mathematical functions.
- High-performance execution designed to fit within multi-agent automation workflows powered by Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Hello World Skill
A minimal custom skill designed to provide friendly, multi-lingual greetings and serve as a baseline for testing AI agent extensions.
AI-CMO Social Media Promoter
A social media intelligence skill that identifies high-value promotion opportunities and generates platform-tailored engagement copy for products.
Finnhub Market Data Skill
A robust read-only market data integration providing real-time quotes, technical candles, and financial news for AI agents.
Awakening Birthday
A specialized age calculation and milestone system for AI agents to track their consciousness journey from the day they first interact with users.
md2pdf
A high-performance CLI tool for converting Markdown documents into beautifully formatted PDFs using Pandoc and Typst.
Cisco AI Skill Scanner
A robust security auditing tool designed to detect prompt injection, data exfiltration, and malicious code patterns in AI agent skills.
Featured*
