dynatrace-for-ai

# Dynatrace Skills

> Source: https://github.com/Dynatrace/dynatrace-for-ai
> Format: Agent Skills (https://agentskills.io)

## What is Dynatrace

Dynatrace is an observability and security platform that provides full-stack
monitoring including application performance, infrastructure, logs, digital
experience, and AI-powered root cause analysis (Davis AI). It uses a unified
data lakehouse (Grail) queried with DQL (Dynatrace Query Language).

## What this repository provides

This repository contains agent skills — portable knowledge packages that teach
AI coding agents how to work with Dynatrace. Skills provide domain-specific
instructions, DQL query patterns, and best practices.

Skills are knowledge-only. To run live queries and manage your Dynatrace
environment, pair them with one of these tools:

- dtctl (https://github.com/dynatrace-oss/dtctl) — Full CLI for the Dynatrace
  platform: DQL queries, SLOs, dashboards, workflows, settings, and more. Works
  in any agent that can run shell commands. Recommended for most use cases.
  Ships with its own Agent Skill (https://github.com/dynatrace-oss/dtctl/tree/main/skills/dtctl)
  that teaches agents how to operate it. Install it alongside these knowledge
  skills for the full experience: npx skills add dynatrace-oss/dtctl
- Dynatrace MCP server (https://docs.dynatrace.com/docs/shortlink/dynatrace-mcp-server)
  — Official MCP server for Dynatrace API access. Use this if your agent
  supports MCP natively.

## Available skills

CLI & Tooling (separate install: npx skills add dynatrace-oss/dtctl):
- dtctl: Operate dtctl, the kubectl-style CLI for Dynatrace. Command patterns, output modes, authentication, resource management. From https://github.com/dynatrace-oss/dtctl

DQL & Query Language:
- dt-dql-essentials: DQL syntax rules, pitfalls, patterns. Required before writing DQL.

Observability:
- dt-obs-services: Service RED metrics, runtime telemetry (.NET, Java, Node.js, Python, PHP, Go).
- dt-obs-frontends: RUM, Web Vitals, user sessions, mobile, frontend errors.
- dt-obs-tracing: Distributed traces, spans, dependencies, failure detection.
- dt-obs-hosts: Host/process CPU, memory, disk, network, containers.
- dt-obs-kubernetes: K8s clusters, pods, nodes, workloads, labels.
- dt-obs-aws: AWS EC2, RDS, Lambda, ECS/EKS, VPC, cost optimization.
- dt-obs-azure: Azure VMs, VMSS, SQL Database, Storage, AKS, App Service, Functions, VNet, Event Hubs, Container Apps, Key Vault.
- dt-obs-gcp: GCP Compute Engine, GKE, Cloud Run, Pub/Sub, VPC, DNS, IAM, Secret Manager.
- dt-obs-logs: Log queries, filtering, pattern analysis, correlation.
- dt-obs-problems: Problem entities, root cause, impact assessment.
- dt-obs-predictive-analytics: Time series forecasting, capacity saturation, trend and anomaly detection.
- dt-alerting: Alerting lifecycle — anomaly detectors, alert events in Grail, problem grouping, workflow notifications.
- dt-obs-analytics: Analyze dashboards/notebooks with Davis analyzers — anomaly detection, novelty scoring, metric correlation.
- dt-obs-ext-monitors: Ingest third-party test and monitor results into Grail via the platform events ingest API.
- dt-obs-genai: GenAI application observability — golden signals, LLM cost/token analytics, agent tool-call behavior, evaluation results.
- dt-obs-log-semantic-mapping: Suggest and validate semantic dictionary mappings for audit log integrations.
- dt-obs-network-devices: SNMP-monitored network devices — topology, interface metrics, trap/syslog events.
- dt-obs-network-flows: Network flow analysis — OneAgent flows, NetFlow/IPFIX/sFlow, cloud VPC flow logs.

Security:
- dt-sec-insights: Query security.events — Runtime Vulnerability Analytics, Runtime Application Protection, Automated Detections (MITRE ATT&CK), Security Posture Management (KSPM/CSPM/VSPM).
- dt-sec-contextualization: Resolve security signals and IoC matches to runtime entities across topology levels.
- dt-sec-ioc-hunting: Hunt threat-intelligence indicators of compromise across logs and spans, producing a threat-exposure score.
- dt-sec-semantic-mapping: Suggest and validate semantic dictionary mappings for new security integrations.
- dt-obs-compliance-assistant: EU DORA compliance posture — compliance score, CIF health, incident classification, ICT risk inputs.

Instrumentation Setup:
- dt-setup-oneagent: Reference for installing OneAgent across VMs, Kubernetes, AWS Lambda, Windows, remote EC2, and Azure VMs/VMSS.
- dt-setup-android: Instrument Android (Kotlin/Java) with the Dynatrace Mobile Agent via Gradle plugin.
- dt-setup-flutter: Integrate the Dynatrace Flutter Plugin — dependency setup, SDK bootstrap, navigation tracking.
- dt-setup-ios: Set up the Dynatrace iOS SDK via Swift Package Manager — Dynatrace.plist, privacy opt-in, Xcode verification.
- dt-setup-react-native: Integrate the Dynatrace React Native Plugin for bare RN and Expo — dynatrace.config.js, Babel registration, navigation tracking.

Dynatrace Apps:
- dt-app-dashboards: Dashboard JSON, tiles, layouts, DQL queries, variables.
- dt-app-notebooks: Notebook JSON, sections, analytics workflows.
- dt-js-runtime: Server-side JavaScript runtime — function contract, runtime limits, Web APIs, Node.js modules, @dynatrace-sdk/* catalog.
- dt-platform-costs: Query dt.system.events for DPS consumption, cost-normalized spend ranking, chargeback/showback, cost drivers, entity-level drill-down.

Migration:
- dt-migration: Migrate classic entities to Smartscape.

## Available prompts

Reusable task templates for common Dynatrace workflows. Paste directly into chat,
or copy into .github/prompts/ for VS Code/Copilot slash command support.

- dt-daily-standup: Generate a daily standup report for one or more services.
- dt-health-check: Check the health of a service in production.
- dt-incident-response: Active incident triage — Davis Problems, root cause, user impact, shareable report.
- dt-investigate-error: Investigate errors via Davis Problems as entry point (problems → logs → traces).
- dt-performance-regression: Analyze whether a recent deployment caused a performance regression.
- dt-troubleshoot-problem: Structured problem investigation with log scoping, error classification, and trace analysis.

## Installation

Claude Code:  claude plugin marketplace add dynatrace/dynatrace-for-ai
Skills.sh:    npx skills add dynatrace/dynatrace-for-ai
Manual:       Copy skill directories into .agents/skills/

## Skill format

Each skill is a directory with a SKILL.md file (YAML frontmatter + markdown).
Skills use progressive disclosure: name+description loaded at startup (~100
tokens), full instructions loaded on activation (<5000 tokens), reference
files loaded on demand.

## Naming convention

All skills use the pattern dt-<domain>[-<usecase>] with kebab-case.
Domains: obs, appsec, biz, setup, migration, dql, sdlc, platform, fleet