A comprehensive bridge between natural language and Prometheus-compatible metrics servers for real-time monitoring and incident triage.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install prom-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 prom-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 prom-query?
prom-query is a specialized tool designed to integrate seamlessly with Openclaw Skills, enabling developers and SREs to interact with Prometheus-compatible metrics servers using natural language. It supports querying metrics, interpreting complex timeseries data, and triaging alerts across platforms like Prometheus, Thanos, Mimir, and VictoriaMetrics through a unified HTTP API.
By translating user intent into precise PromQL, this skill simplifies observability tasks and provides actionable insights into system health without requiring the user to be an expert in query syntax. It is an essential component for any automated monitoring workflow powered by Openclaw Skills, ensuring that infrastructure data is always accessible and interpretable.
prom-query Use Cases
- Triage firing alerts and investigate root causes during production incidents.
- Calculate error rates, latency percentiles (P99/P50), and resource utilization via simple text commands.
- Monitor Kubernetes pod health, identify CrashLoopBackOff states, and check node conditions.
- Explore available metrics and search for specific patterns within high-cardinality environments.
- Predict disk exhaustion or capacity trends using linear regression queries to prevent downtime.
How prom-query Works
- The user provides a natural language request or a specific monitoring command to the agent.
- The skill translates the request into a valid PromQL query or a specific API call for alerts, targets, or rules.
- It executes the command against the configured Prometheus-compatible server using the provided URL and authentication token.
- Results are automatically processed, including downsampling for large datasets and the generation of statistical summaries (min, max, average).
- The skill returns a structured interpretation of the metrics, highlighting trends, spikes, or step changes for the user.
prom-query Setup
To integrate this skill into your workflow, ensure the following environment variables are configured. This setup allows Openclaw Skills to securely communicate with your metrics provider.
export PROMETHEUS_URL="https://your-prometheus-instance.com"
# Optional: Set a bearer token for authenticated instances
export PROMETHEUS_TOKEN="your-access-token"
prom-query Data Schema & Taxonomy
The prom-query skill interacts with the Prometheus HTTP API and returns structured metrics data optimized for agent interpretation. Data is organized as follows:
| Field | Description |
|---|---|
min / max / avg |
Statistical summaries of all values within the retrieved timeseries. |
first / last |
The start and end values of the range, used to identify growth or decline trends. |
pointCount |
The total number of data points returned in the query result. |
downsampled |
A boolean flag indicating if the step was increased to limit data volume for better performance. |
activeAt |
Timestamp for alerts indicating when the firing state began. |
prom-query Advanced Features
- Automatic downsampling of range queries exceeding 500 data points to maintain context window efficiency.
- Discord v2 Delivery Mode support, providing compact summaries and interactive quick actions in supported chat environments.
- Cross-platform compatibility with Prometheus, Thanos, Mimir, and VictoriaMetrics.
- Regex-based metric exploration for rapid discovery of unknown service metrics.
- Read-only safety ensures that Openclaw Skills can query data without the risk of modifying server configurations or rules.
SKILL.md
Loading
Related Openclaw Skills
PagerDuty Incident Triage
Automate your SRE and incident response workflows with AI-powered PagerDuty triage and management.
Meeting Autopilot
An operational AI skill that converts meeting transcripts into structured outputs including action items, decisions, and ready-to-send email drafts.
Log Dive
A unified log search engine that translates natural language queries into LogQL, ES DSL, or CloudWatch filter patterns across multiple backends.
kube-medic
A powerful Kubernetes diagnostic toolkit for instant AI-driven cluster triage and incident response using standard kubectl commands.
rug-checker
An automated on-chain security auditor for Solana tokens that detects rug-pull risks without requiring API keys or wallet connections.
Vibe Check
Vibe Check is an automated auditing tool that detects and scores lazy AI-generated coding patterns to ensure your codebase maintains high human-level standards.
METADATA
Featured*
