Proposal: Integrating Dokploy MCP – Real-World Ecommerce Use Case, Automation & AI Agents

[andradehenrique]

Hi Dokploy team! 👋

My name is Henrique (@andradehenrique), and I’m a developer at Thepes. We’re currently migrating a large ecommerce platform, made up of several microservices, from GCP to a VPS architecture, with Dokploy as a central piece for automation and orchestration.

During this integration process, I created a Dokploy MCP (Model Context Protocol), which is available here:
👉 https://github.com/andradehenrique/dokploy-mcp
The initial goal was to generate tools directly from the Dokploy Swagger API, making it much easier to integrate with AI agents and build advanced automation workflows.

Beyond the ecommerce migration, our company also develops custom AI agents for business automation and customer support. We see great opportunities for the Dokploy MCP in scenarios such as automation via AI, n8n workflows, and other contexts where Dokploy actions need to be triggered programmatically.

Proposal:
I’d like to know if the Dokploy team would be interested in officially adopting this MCP (for example, moving the repository to the Dokploy organization). I’m fully committed to maintaining and evolving this project, as it will be actively used in our company and I believe it can bring significant value to the wider community.

If you find this interesting, I’m happy to discuss how we could best structure this collaboration and ensure a smooth integration.

Thank you for your great work with Dokploy!
Best regards,
Henrique

[Siumauricio]

Hi, thanks for your interest, I think someone also did a MCP btw https://github.com/apple-techie/dokploy-mcp

I don't have much experience with MCP, could you give me an important case that could be ideal to integrate this?

I can be open to move that repository to the dokploy org and add it as a maintainer for that specific project.

[andradehenrique]

Hi @Siumauricio,

Thank you for your interest! As requested, here are some important real-world use cases for the Dokploy MCP, especially when combined with other Model Context Protocol (MCP) servers from the official ecosystem:

---

1. Orchestration with AI Agents & Automation Platforms (n8n, Zapier, Chatbots, Claude, GPT, etc.)
With a Dokploy MCP, AI agents or automation flows (e.g., n8n, Zapier, chatbots) can deploy, monitor, or scale infrastructure automatically—triggered by events, prompts, or incidents.
Example: An AI agent detects a critical alert from Sentry (using the Sentry MCP), and through the Dokploy MCP, automatically restarts a service or scales resources, possibly after discussing the incident with a human via Slack MCP.
Typical integrations: Sentry MCP (alerts), Slack MCP (notifications), GitHub MCP (open issues on failed deploys), Filesystem MCP (log/file handling).

---

2. Smart, Agnostic CI/CD Pipelines
The Dokploy MCP provides a unified interface so that CI/CD pipelines (GitHub Actions, GitLab CI, Jenkins, etc.) can trigger Dokploy deployments in a language-agnostic way.
Example: After a successful build, the pipeline calls the Dokploy MCP for deployment, sends notifications to teams via Slack MCP, and registers deployment status in GitHub MCP or Notion MCP.

---

3. Self-Healing Infrastructure & Proactive Incident Response
By integrating with monitoring MCPs (Sentry, Datadog, Grafana), you can enable automated workflows that react to incidents without human intervention.
Example: If a container failure is detected (by Grafana MCP), an agent can use the Dokploy MCP to restart the affected service and update status via Slack MCP.

---

4. Programmatic Multi-Tenant SaaS Management
For SaaS platforms with many tenants, the Dokploy MCP allows programmatically creating, updating, or scaling tenant environments via API or AI agents—accelerating onboarding and maintenance.
Example: When a new client signs up, a CRM or database integration (e.g., Notion MCP, PostgreSQL MCP) triggers the Dokploy MCP to spin up an isolated environment automatically.

---

Synergy with Other MCP Servers

By combining the Dokploy MCP with other servers from the [official MCP repository](https://github.com/modelcontextprotocol/servers), we enable rich, modular, and auditable workflows. Relevant servers include:

• GitHub MCP – Automate issues, releases, merges, and deployment feedback.
• Slack MCP – Smart notifications for ops and dev teams.
• Sentry MCP – Automatic remediation on application errors.
• Filesystem MCP – Handle configuration files and logs.
• PostgreSQL MCP – Data-driven workflows for onboarding, migrations, etc.
• Grafana MCP – Monitoring and automation based on live metrics.

---

These scenarios demonstrate how the Dokploy MCP unlocks secure, programmable, and scalable automation for DevOps, SaaS, and AI-driven environments.

[andradehenrique]

As a quick update, the current implementation already covers all project and application endpoints. My goal is to achieve 100% coverage of the official Dokploy Swagger API in a short timeframe, since the codebase for tool creation is already set up and modular.

All tools are fully typed and validated using zod, strictly following the Swagger definitions. This ensures a robust and developer-friendly experience, with clear type safety and reliable input validation.

[Siumauricio]

thank you for the detailed explanation

I think it's fine, I create a repository called mcp, you can see the invitation here https://github.com/Dokploy/mcp/invitations, you should have received an invitation in your email as well.

[andradehenrique]

Thank you very much for your quick response and for creating the repository!

I’ve just accepted the invitation and will start migrating the code as soon as possible. I’m excited to contribute and keep the Dokploy MCP up to date and aligned with the official Swagger spec.

If you have any suggestions, priorities, or guidelines for the repo structure, feel free to share them—I’m happy to collaborate closely to make this project valuable for the community.

Thanks again for the trust and opportunity!

[andradehenrique]

Hi @Siumauricio,

I've successfully migrated the repository to the new organization—thanks again for the support!

I have a quick question regarding the npm package:
Currently, the Dokploy MCP is published under my personal scope as [@ahdev/dokploy-mcp](https://www.npmjs.com/package/@ahdev/dokploy-mcp).
Would you prefer that I keep publishing updates there, or should we consider publishing it under the official Dokploy organization scope (for example, @dokploy/mcp), similar to how the CLI package is published ([@dokploy/cli](https://www.npmjs.com/package/@dokploy/cli))?