signoz - SKILL.md Agent Skill

name: signoz description: Query SigNoz observability platform for logs, metrics, and traces. Use when user asks to check logs, query metrics, analyze traces, monitor services, or investigate issues in SigNoz. Supports filtering by service, time range, and custom queries. Triggers include 'check logs in signoz', 'query metrics', 'look at traces', 'what's happening with service X', 'show me errors', 'monitoring dashboard'.

SigNoz Observability

Query logs, metrics, and traces from SigNoz API.

Quick Start

# Set environment variables
export SIGNOZ_API_KEY=your-key
export SIGNOZ_URL=https://logs.s4l.link

# Query recent logs
signoz logs --service=magento --last=1h

# Check metrics
signoz metrics cpu_usage --start=24h

# View traces
signoz traces --service=api --limit=20

# List all services
signoz services

CLI Tool

Use scripts/signoz-cli.py for all queries:

python3 scripts/signoz-cli.py logs --service=<name> [options]
python3 scripts/signoz-cli.py metrics <metric_name> [options]
python3 scripts/signoz-cli.py traces [options]
python3 scripts/signoz-cli.py services

Logs Command

signoz logs --service=magento --query="error" --start=1h --limit=50

Options:

--service, -s — Filter by service name
--query, -q — Search query string
--start — Start time (e.g., 1h, 30m, 1d, or timestamp)
--end — End time (timestamp)
--limit, -l — Max results (default: 100)

Metrics Command

signoz metrics cpu_usage --start=24h --step=300 --aggregation=avg

Options:

metric — Metric name (required)
--start — Start time (e.g., 1h, 24h, 7d)
--end — End time
--step — Step interval in seconds (default: 60)
--aggregation, -a — Aggregation type: avg, sum, min, max, count

Traces Command

signoz traces --service=api --operation=GET /orders --start=30m

Options:

--service, -s — Filter by service
--operation, -o — Filter by operation name
--start — Start time
--end — End time
--limit, -l — Max results (default: 50)

Services Command

signoz services

Lists all services reporting to SigNoz.

Common Use Cases

Debug Production Issues

# Check recent errors
signoz logs --service=magento --query="error" --start=15m

# Look at slow traces
signoz traces --service=api --start=1h | head -20

# Check resource usage during incident
signoz metrics memory_usage --start=2h --step=60

Monitor Service Health

# What's running?
signoz services

# Recent activity by service
signoz logs --service=payment-gateway --start=1h --limit=10

# Error rate
signoz metrics http_requests_total --start=24h --aggregation=count

Performance Analysis

# Trace latency distribution
signoz traces --service=api --start=1h

# Resource trends
signoz metrics cpu_usage --start=7d --step=3600

Time Ranges

Use relative time for convenience:

5m — 5 minutes ago
30m — 30 minutes ago
1h — 1 hour ago
6h — 6 hours ago
24h / 1d — 1 day ago
7d — 7 days ago

Or use millisecond timestamps for precise ranges.

Environment Setup

Required environment variables:

export SIGNOZ_API_KEY=FS3pXDYQsAbBUAvJeOMxiPbXcVJSRafe8rpKjb12cx8=
export SIGNOZ_URL=https://logs.s4l.link

Add to ~/.bashrc or ~/.zshrc for persistence.

API Reference

For detailed API documentation, see references/api-reference.md.

Tips

Start with broad time ranges, then narrow down
Use --limit to avoid overwhelming output
Filter by service first to reduce noise
Combine with grep, jq, or head for further processing
For dashboards, use the SigNoz web UI; use CLI for quick checks

Troubleshooting

"API key required" error → Set SIGNOZ_API_KEY environment variable

Empty results → Check time range (might be too narrow) → Verify service name spelling → Try without filters first

Connection errors → Check SIGNOZ_URL is correct → Verify network access to SigNoz instance