AI inside the panel

Manage your server
with words

The AI assistant knows your infrastructure: it deploys, reads logs and metrics, runs commands and SQL. And the MCP server connects the same tools to your favorite AI — with your own keys, scopes and audit.

Try for free All features

Ask — and it gets done

The assistant doesn't just answer — it calls real panel tools and shows every step.

Deploykin Assistant

Why is shop.example.com returning 502?

Checking the site's logs and metrics…

tail_log nginx-error server_metrics web-01 read_configuration fpm-pool

PHP-FPM hit its limit: pm.max_children = 16, all workers busy. Raised it to 32 and restarted the pool — the site returns 200 again.

OK. Deploy the latest commit to production and back up the database.

run_backup shop-db deploy_website production

Done: backup shop-db (245 MB) uploaded and encrypted. Zero-downtime deploy of commit a1b9c4e finished in 14 s.

Type what you need done…

Dangerous and state-changing actions can require confirmation, and every tool call is recorded in the audit log.

MCP-server

Your AI — with access to your infrastructure

Deploykin runs a Model Context Protocol server. Connect Claude, Cursor or any MCP client — and it gets the same tools as the built-in assistant.

One endpoint — 30+ panel tools
MCP standard: works with any compatible client
Scoped tokens: grant exactly the rights that are needed
Every tool call is written to the audit log

~/.cursor/mcp.json

{
  "mcpServers": {
    "deploykin": {
      "url": "https://app.deploykin.com/mcp",
      "headers": {
        "Authorization": "Bearer dk_mcp_••••4f2a"
      }
    }
  }
}

30+ tools out of the box

Everything you do by hand in the panel, the assistant does with a tool call.

Deploy

deploy_website

wait_for_task

restart_task

Databases

db_query

db_dump

db_execute

Commands

run_command

run_artisan

run_wp_cli

Metrics

server_metrics

get_server

Logs

tail_log

list_logs

Backups

run_backup

list_backups

SSL

renew_ssl

read_env

Firewall

create_firewall_rule

list_firewall_rules

Deploykin AI out of the box — or your key and your model

On the Pro plan the assistant runs on Deploykin AI — no keys, no setup. On Business, connect your own key from your favorite provider and pick the model: tokens are billed directly to them, with no markup from us. Conversation history is saved in the panel.

Anthropic

Claude

OpenAI

GPT

Google

Gemini

DeepSeek

Access control

Scoped tokens and full audit

Issue a separate token for each client or script. Limit rights to the necessary minimum and revoke in one click. Every tool call is logged.

Read and write scopes for each area
Restrict by servers and projects
Instant revocation and token expiry
Audit log: who, when, which tool, result

dk_mcp_••••4f2a

active

Scopes

deploy:write db:read logs:read metrics:read backup:write

Audit log

deploy_website production just now

run_backup shop-db 1 min ago

db_query shop-db 3 min ago

create_firewall_rule web-01 10 min ago

Deploykin bot

Deploy main to prod and clean up old releases

deploy_website

Deploy main → prod started…

SUDO Approval needed

rm -rf releases/old-*

✓ Approve

✕ Deny

Telegram bot

The same assistant — right in Telegram

Text the bot in plain words — and it deploys, reads logs, runs a command or SQL. The same assistant, same tools and same limits as in the panel. The reply arrives in real time.

Approve dangerous actions with a button right in the chat — marked SUDO
Command arguments are frozen: the model can't swap them after you approve
Alerts land here too: expiring SSL, a failed deploy, an unreachable server, a full disk
Connect with a single link from the panel settings

Available on the Pro and Business plans.

Hand the AI the keys to your routine

The assistant and MCP server are available on the Pro and Business plans. Open the chat in the panel — and start managing your infrastructure with words.

Try for free View pricing

Manage your serverwith words