How to Use the Heroku (PaaS) MCP in Pydantic AI

Q: How do I attach Heroku (PaaS) to Pydantic AI?

You initialize an MCPToolset with your HTTP endpoint. Pass that toolset array directly into your Agent constructor. The old MCPServerHTTP class is deprecated.

Q: Does Pydantic AI validate Heroku (PaaS) tool responses?

Yes, it does. Every time the agent runs a tool like listaddons, the framework checks the output against its internal schema. You get strict type safety for all your infrastructure queries.

Q: How do I check if my dynos are crashing?

The agent calls listdynos. It tracks the exact state of each container based on the internal slug runner engine. It knows immediately if a process is starting, idle, or crashed.

Q: What are the privacy risks of reading Heroku (PaaS) variables?

This server exposes the listconfigvars tool, which retrieves highly confidential database tokens and SendGrid passwords. Pydantic AI validates this data, but those secrets are processed by whatever LLM provider you configured for the agent.

Type-safe Heroku management for Pydantic AI via this MCP Server.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Heroku (PaaS) MCP to Pydantic AI

Create your Vinkius account to connect Heroku (PaaS) to Pydantic AI and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup Heroku (PaaS) with Pydantic AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Type-Safe Heroku (PaaS) MCP Server Operations

Your agent needs predictable data. When it calls `get_app_info`, this MCP Server returns the exact routing URLs and slug sizes. Pydantic AI validates that payload against your models at runtime. If the API format shifts, the agent fails loudly instead of hallucinating. The same applies to fleet discovery. The AI client executes `list_apps` to discover git repository targets. You get a clean, validated array of applications. No silent corruption and no missing fields.

Precision Dyno Control

Stop waking up to fix memory leaks. Your agent monitors the telemetry via `list_dynos` to check if a container is crashed or idle. If it spots a stuck process, it fires `restart_specific_dyno` to unstick that isolated instance. Sometimes the whole application stalls out. The agent runs `restart_all_dynos` to perform a hard reboot across all containers. This clears the memory leak and restores request processing instantly.

Safe Migrations and Provisioning

You can't run database migrations with active web traffic. Your agent handles this by executing `toggle_maintenance_mode` before touching anything. It blocks new connections so you avoid corrupted states. When you need a fresh environment, the agent uses `create_app`. It provisions a new structural container. If you're done with a test environment, it runs `delete_app` to permanently wipe it from the servers.

Setup guide

Set up Heroku (PaaS) MCP in Pydantic AI

Prerequisites

Python 3.10+ installed
pydantic-ai-slim[fastmcp] package
Active Vinkius subscription with a valid endpoint token

1

Install Pydantic AI with FastMCP
Run pip install "pydantic-ai-slim[fastmcp]". The FastMCP toolset replaces the deprecated MCPServerHTTP class with full protocol support.
2

Configure the FastMCPToolset
Pass a JSON-style config dict to FastMCPToolset with your Vinkius URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. Supports Streamable HTTP, SSE, and Stdio transports.
3

Create and run your agent
Pass the toolset to Agent(toolsets=[toolset]) and call agent.run(). Swap openai:gpt-4o for any supported model — Anthropic, Google, Mistral, or Groq.

agent.py

from pydantic_ai import Agent
from pydantic_ai.toolsets.fastmcp import FastMCPToolset

toolset = FastMCPToolset({
    "mcpServers": {
        "heroku-paas-mcp": {
            "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
        }
    }
})

agent = Agent(
    "openai:gpt-4o",
    toolsets=[toolset],
    system_prompt="You have access to Heroku (PaaS) tools.",
)

result = await agent.run("List recent Heroku (PaaS) transactions")
print(result.output)

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Heroku. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect Heroku (PaaS) now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about Heroku (PaaS) MCP in Pydantic AI

You initialize an `MCPToolset` with your HTTP endpoint. Pass that toolset array directly into your Agent constructor. The old `MCPServerHTTP` class is deprecated.

Yes, it does. Every time the agent runs a tool like `list_addons`, the framework checks the output against its internal schema. You get strict type safety for all your infrastructure queries.

You certainly can. Pydantic AI is model-agnostic. As long as the model supports tool calling, it can execute `restart_all_dynos` or `list_apps` using this MCP Server.

The agent calls `list_dynos`. It tracks the exact state of each container based on the internal slug runner engine. It knows immediately if a process is starting, idle, or crashed.

This server exposes the `list_config_vars` tool, which retrieves highly confidential database tokens and SendGrid passwords. Pydantic AI validates this data, but those secrets are processed by whatever LLM provider you configured for the agent.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript