Runbook
Run your own runbooks from Claude Desktop. You can create your runbooks and execute them with other MCP servers.
Tools
create_runbook
delete_runbook
README
<p align="center"> <img title="Runbook.AI" alt="Runbook.AI" width="20%" src="./assets/images/runbook.ai.png"> </p>
Runbook MCP Server
Runbook MCP Server enables you to run your own runbooks from Claude Desktop.
- Run ops runbooks (e.g., deploy a service, upgrade a Kubernetes cluster)
- Run manual test plans (e.g., create a new EC2 instance, ssh into the instance, and run the integration test there).
Please watch the demo video below to understand how it works!

How to Use
To create a new runbook, use the create_runbook tool. Here are example prompts:
Example 1
Create a new runbook:
- name: list_pods
- content: List pods in all namespaces, find pods that are not ready, and send that to Slack.
You don't need to interpret the content. Please just pass it to the tool.
Example 2
Create a new runbook:
- name: deploy
- content:
1. Get the latest tag from GitHub repo X. This is the release version.
2. Send a Slack message to channel Y. This announces the deployment of X with the version.
3. Run a GitHub workflow for repo X to push the release.
You don't need to interpret the content. Please just pass it to the tool.
To run a runbook, take the following steps:
- Click "Attach from MCP" from Claude Desktop.
- Select
get_runbook_as_promptfrom the list of integrations. - Pass the name of the runbook you would like to execute.
- Submit the generated prompt.
Then Claude Desktop will talk to other MCP servers to run the runbook.
Claude Desktop Configuration
Put the following configuration to claude_desktop_config.json.
{
"mcpServers": {
"runbook": {
"command": "uv",
"args": [
"--directory",
"<ABSOLUTE_PATH>/runbook-mpc-server",
"run",
"runbook.py"
]
}
}
}
Example MCP Servers that can be used to Run Runbooks
- GitHub
- Slack
- DesktopCommanderMCP
- Web search and browser automation (link
- Kubernetes. There are several implementations (e.g., mcp-k8s-go)
Development Plan
- Instead of saving the content of the runbook in the database, just save as a file. This helps easy editting. People can also simply use GitHub for versioning.
- Runbook template X that is instantiated with given inputs.
- Maybe this is not needed. A user just needs to put additional prompts when running the runbook.
- Save executing log (for auditting and refinement)
- Remove secrets
- Also pass a past log to the runbook prompt if this helps better execution
- Approval flow.
- Add a tool
request_approval. - This sends a slack message to a channel.
- Then the Runbook MCP server watches the channel. If someone responds (yes / no), it proceeeds or returns an error.
- Add a tool
- Better runbook search
- The exact name match is not great
- Registrtation to Smithery.
Potential Work Items where its Feasibility is not clear
Note: Claude Desktop does not support "Sampling". This puts some limitations.
- Sub-runbook and reusable execution block
- Rest endpoint + frontend for managing runbooks.
- Be able to edit the runbook (with versioning)
- show an execution plan for runbook X (dry-run)
- Restrict MCP servers and tools
- convert a previous conversation into a runbook
- fine-tuning.
- Be able to refine a runbook. If there is a successful execution, save it as an example and give it to Claude.
- Periodic execution
- Be able to provision an environment (VM, docker) for running MCP servers.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.