The Datadog Observability skill empowers Claude to interact directly with Datadog's full-stack monitoring suite. It provides seamless access to metrics, APM traces, logs, and infrastructure data, allowing for efficient system health monitoring and rapid incident resolution while optimizing context usage for AI-driven operations.
Use Cases
Real-time Incident Troubleshooting: Rapidly search through distributed traces and logs to pinpoint the exact cause of application errors or latency spikes during active production incidents.
Infrastructure & Container Health Checks: Retrieve host-level metrics and list active containers to ensure infrastructure stability and monitor resource allocation efficiency.
Automated Alert Management: List existing monitors, check their current status, and programmatically mute or create alerts to streamline SRE and DevOps workflows.
Performance Analysis & Optimization: Access APM service summaries and query metric timeseries to identify long-term performance trends and optimize application bottlenecks.
name
datadog-observability
description
Datadog monitoring, APM, logs, and infrastructure observability
version
1.0.0
allowed-tools
[Bash, Read, WebFetch]
Datadog Observability Skill
Overview
Provides 90%+ context savings vs raw Datadog API integration. Multi-service support with progressive disclosure by observability category.
Requirements
Datadog API configured
Environment variables:
DD_API_KEY (required): Datadog API key
DD_APP_KEY (required): Datadog application key
DD_SITE (optional): Datadog site (default: datadoghq.com)
Tools (Progressive Disclosure)
Metrics Operations
Tool
Description
Confirmation
query-metrics
Query metric timeseries
No
list-metrics
List available metrics
No
post-metrics
Submit custom metrics
Yes
APM/Traces Operations
Tool
Description
Confirmation
list-services
List APM services
No
service-summary
Get service performance summary
No
search-traces
Search distributed traces
No
Logs Operations
Tool
Description
Confirmation
search-logs
Search log entries
No
log-indexes
List log indexes
No
log-pipelines
View log processing pipelines
No
Monitors/Alerts Operations
Tool
Description
Confirmation
list-monitors
List monitors
No
monitor-status
Get monitor status
No
create-monitor
Create new monitor
Yes
mute-monitor
Mute monitor alerts
Yes
Infrastructure Operations
Tool
Description
Confirmation
list-hosts
List infrastructure hosts
No
host-metrics
Get host-level metrics
No
list-containers
List containers
No
Quick Reference
# Query metrics
curl -X POST "https://api.${DD_SITE}/api/v1/query" \
-H "DD-API-KEY: ${DD_API_KEY}" \
-H "DD-APPLICATION-KEY: ${DD_APP_KEY}" \
-d '{"query":"avg:system.cpu.user{*}"}'
# Search logs
curl -X POST "https://api.${DD_SITE}/api/v2/logs/events/search" \
-H "DD-API-KEY: ${DD_API_KEY}" \
-H "DD-APPLICATION-KEY: ${DD_APP_KEY}" \
-d '{"filter":{"query":"service:web","from":"now-1h","to":"now"}}'
# List monitors
curl -X GET "https://api.${DD_SITE}/api/v1/monitor" \
-H "DD-API-KEY: ${DD_API_KEY}" \
-H "DD-APPLICATION-KEY: ${DD_APP_KEY}"
# Get service summary
curl -X GET "https://api.${DD_SITE}/api/v1/apm/service/{service_name}" \
-H "DD-API-KEY: ${DD_API_KEY}" \
-H "DD-APPLICATION-KEY: ${DD_APP_KEY}"
Configuration
DD_API_KEY: Datadog API key (required)
DD_APP_KEY: Datadog application key (required)
DD_SITE: Datadog site domain (optional, default: datadoghq.com)
US1: datadoghq.com
US3: us3.datadoghq.com
US5: us5.datadoghq.com
EU: datadoghq.eu
AP1: ap1.datadoghq.com
Security
⚠️ Never hardcode API keys or application keys
⚠️ Use environment variables or secret management
⚠️ Monitor mutations (create, mute) require confirmation
⚠️ Never expose DD_API_KEY or DD_APP_KEY in logs or responses
Agent Integration
devops (primary): Infrastructure monitoring and SRE
incident-responder (primary): Crisis management and troubleshooting
