cx-skills

# Coralogix Skills

> Source: https://github.com/coralogix/cx-skills
> Format: Agent Skills (https://agentskills.io)

## What is Coralogix

Coralogix is an observability platform for logs, metrics, traces, and security
data. It supports OpenTelemetry-native collection, real-time analytics,
dashboards, alerting, APM, infrastructure monitoring, and archive-first log
management.

## What this repository provides

This repository contains agent skills: portable knowledge packages that teach
AI coding agents how to work with Coralogix and OpenTelemetry. Skills provide
domain-specific instructions, configuration patterns, troubleshooting flows,
query examples, and best practices.

Skills are knowledge-only. To send telemetry, query data, or manage your
Coralogix environment, pair them with the relevant Coralogix integrations,
OpenTelemetry Collector distributions, SDKs, APIs, or platform tooling.

## Available skills

OpenTelemetry:
- [opentelemetry-collector](skills/opentelemetry/opentelemetry-collector/SKILL.md):
  Deploy, configure, and troubleshoot the OpenTelemetry Collector for Coralogix.
  Covers the `coralogix` exporter, `domain:` versus `endpoint:`, processor
  chains, agent to cluster-collector to gateway topology, the `otel-integration`
  Helm chart, ECS EC2/Fargate, standalone installs, Docker, and the universal
  installer.
- [opentelemetry-instrumentation](skills/opentelemetry/opentelemetry-instrumentation/SKILL.md):
  Instrument Java, Python, Node.js, .NET, and Go applications with OpenTelemetry
  SDKs to send traces, metrics, and logs to Coralogix. Covers OTLP exporter env
  vars, Coralogix auth headers, `cx.application.name`/`cx.subsystem.name` resource
  attributes, APM transaction samplers, Kubernetes Operator injection, and
  debugging missing telemetry from the SDK side.
- [opentelemetry-ottl](skills/opentelemetry/opentelemetry-ottl/SKILL.md):
  Write and debug OpenTelemetry Transformation Language (OTTL) statements for
  transform processors, filter processors, and routing connectors. Covers
  context selection, path expressions, `error_mode`, cardinality reduction, JSON
  body parsing, PII redaction, and span naming.

## Installation

Agent Skills CLI: `npx skills add coralogix/cx-skills`
Claude Code:      `claude plugin marketplace add coralogix/cx-skills`
                  then `claude plugin install coralogix@cx-skills`
Cursor GitHub:    Add `https://github.com/coralogix/cx-skills` from Cursor
                  Settings -> Rules, Skills, Subagents
Cursor CLI:       `git clone https://github.com/coralogix/cx-skills ~/.cursor/skills/cx-skills`

All install paths load the same `skills/**/SKILL.md` directories. The
`.codex-plugin/`, `.claude-plugin/`, and `.cursor-plugin/` files are discovery
metadata only.

## Skill format

Each skill is a directory with a `SKILL.md` file containing YAML frontmatter and
markdown instructions. Skills use progressive disclosure: the name and
description are loaded first, then full instructions and references are loaded
when the task activates the skill.

## Naming convention

Skills use descriptive kebab-case names. OpenTelemetry skills use the
`opentelemetry-<domain>` pattern, such as `opentelemetry-collector`,
`opentelemetry-instrumentation`, and `opentelemetry-ottl`.