mcpField
C
coolify

coolify

github.com/oakossCloud & DevOps20362 installsUpdated 10w ago
Repository

About coolify

coolify is a skill for deploying and managing applications, databases, and services on self-hosted Coolify infrastructure. It guides API operations and deployment workflows, including build configuration and CI/CD automation.

Key features

  1. 01Manage applications through the REST API
  2. 02Create and update application settings
  3. 03Trigger deployments with API calls or webhooks
  4. 04Configure Nixpacks, Dockerfile, and Compose builds
  5. 05Set domains, environment variables, and health checks

Use cases

  1. 01Deploy applications to self-hosted infrastructure
  2. 02Automate deployments from CI pipelines
  3. 03Inspect and manage Coolify resources
  4. 04Configure multi-service Compose deployments

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/oakoss/agent-skills --skill coolify

After installing, type/coolifyin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

references/
api-patterns.md
build-packs.md
deployment-strategies.md

Details

From the project README

Coolify

Overview

Coolify is a self-hosted PaaS (Vercel/Netlify alternative) that deploys applications, databases, and services on your own infrastructure. It provides a REST API for programmatic control, supports multiple build packs, and uses Traefik for reverse proxy and SSL.

When to use: Deploying applications to self-hosted infrastructure, managing databases and services through API, automating deployments via CI/CD, self-hosting without vendor lock-in.

When NOT to use: Initial Coolify server installation (use the official install script via browser), managing Coolify's own infrastructure (Postgres, Redis — managed internally), or when a managed platform (Vercel, Railway) meets your needs without self-hosting requirements.

Quick Reference

PatternApproachKey Points
API authAuthorization: Bearer <token> headerToken created in dashboard, scoped permissions
List applicationsGET /api/v1/applicationsReturns all apps with status and config
Trigger deployGET /api/v1/deploy?uuid=<app>&force=trueOr use per-app webhook URL
Create applicationPOST /api/v1/applications/publicRequires project_uuid, server_uuid, env_name
Update applicationPATCH /api/v1/applications/<uuid>Accepts build, health check, domain, limit fields
Auto-deployEnable in Advanced > GeneralDeploys on every push to configured branch
Webhook deploycurl GET <webhook_url> -H "Authorization: Bearer ..."Controlled deploys from CI pipelines
GitHub Actions deployBuild image, push to registry, trigger webhookFull control over build + test before deploy
Nixpacks buildAuto-detected from source codeZero-config for supported languages
Dockerfile buildSet build pack to dockerfileFull control over build process
Compose builddocker-compose.yaml as source of truthMulti-service stacks with Coolify labels
Environment variablesAPI or dashboard per-environmentInterpolated in Compose via ${VAR} syntax
Domain routingfqdn field or Traefik labels in ComposeAutomatic SSL via Let's Encrypt
Health checksConfigurable path, interval, retries via APIRequired for running:healthy status

Common Mistakes

MistakeCorrect Pattern
API returns "Unauthenticated"Add Accept: application/json header alongside Bearer token
Can't read sensitive fields (auto-deploy, etc.)Token needs view:sensitive or * permission scope
Auto-deploy not triggeringVerify GitHub App is connected and auto-deploy enabled in Advanced
Using API token without deploy permissionToken needs explicit deploy permission to trigger deployments
Health check fails on Compose servicesAdd exclude_from_hc: true label for non-HTTP services
Webhook deploy without auth headerAlways include Authorization: Bearer <token> with webhook calls
Environment variables not available in buildUse build arguments in Advanced menu, not runtime env vars
Compose services can't communicateUse service names as hostnames within the same stack network

Delegation

  • API exploration: Use Explore agent to discover existing Coolify resources
  • Dockerfile optimization: Use Task agent to review Dockerfiles for Coolify deployment
  • CI/CD pipeline design: Use Plan agent for deployment workflow strategy

If the docker skill is available, delegate Dockerfile authoring, multi-stage builds, and image optimization to it. If the github-actions skill is available, delegate workflow syntax and CI pipeline patterns to it. If the ci-cd-architecture skill is available, delegate deployment strategy and environment promotion to it.

References

Comments

Markdown supported