> For the complete documentation index, see [llms.txt](https://docs.kinesis.network/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.kinesis.network/mcp.md).
# MCP & agents
Deploy, inspect and operate Kinesis from your agent.
Connect a compatible agent to Kinesis over the Model Context Protocol. Use natural-language workflows backed by tools for projects, apps, repositories, grids, servers and billing.
## Choose your app
Select the app you use for its setup and sign-in steps.
{% hint style="info" %}
Kinesis directory submissions for Claude, ChatGPT and Cursor are awaiting approval. These manual setups are available now.
{% endhint %}
## Start with a connection check
After connecting, send this prompt in your agent:
```
Use Kinesis to show my identity and list the organizations I can access.
Then ask which organization to use and list its projects. Do not change anything.
```
Then try a [deployment or troubleshooting workflow](/mcp/workflows.md), or explore the [tool reference](/mcp/tools.md).
## What your agent can do
* Deploy from an image, a template, a connected GitHub repository or a clone of a public repository; copy environment variables with `envFrom`.
* Read configuration, status, runtime logs, usage and image-build output; check and roll back to an earlier stored build image.
* Check placement, move apps, manage logs, operate grids and dedicated servers, attach machines and manage SSH keys.
* Discover and create organizations; manage members, invitations, settings and the audit log within your role and scopes. Add organization members to private resources and handle notifications.
* Inspect invoices and ledger history, manage saved payment methods and node protection, and redeem vouchers. Adding credits and configuring auto-reload happen in the portal.
Serverless placement is CPU-only; GPU apps need a grid. Logs are raw output and may contain sensitive data.
Actions run as the connected person in each organization, with their role there. Organization-level actions take its slug as `org`. A connection can be limited to chosen organizations; a project-limited key cannot run organization-wide administration. Spending and destructive operations have explicit requirements; the tool reference describes them.
Use `list_orgs` (or `account` → `list_orgs` in compact mode) to choose an organization, then `read_docs` with `topic: "organizations"` for the rules. [Organization actions](/mcp/tools/organizations.md) and [member actions](/mcp/tools/members.md) include full schemas. Older OAuth connections need approval for the new organization scopes.
For integrations that require credential restrictions, use the [OpenAI profile](/mcp/openai-profile.md): `?profile=openai` exposes 94 individual tools and restricts credential inputs and outputs. The ordinary compact connection remains available.
## Connection settings
Use the [shared connection options](/mcp/connect.md#connection-options) to choose individual or compact tools, load selected groups, or expose read-only tools. The [custom agent guide](/mcp/connect/custom-agent.md) gives the server URL and generic authentication steps.
[OAuth details for client developers](/mcp/oauth.md) · [Authentication and scopes](/api/authentication.md)
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.
Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:
```
GET https://docs.kinesis.network/mcp.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
Claude Code
Codex
Cursor
Custom agent