agentcontrolplane
starsforks
How to install agentcontrolplane?
humanlayer/agentcontrolplane
$git clone https://github.com/humanlayer/agentcontrolplane</div>
README.md
| 1 | <div align="center"> |
| 2 | |
| 3 | <h1>Agent Control Plane (ACP)</h1> |
| 4 | |
| 5 | </div> |
| 6 | |
| 7 | ACP (Agent Control Plane) is a cloud-native orchestrator for AI Agents built on Kubernetes. It supports [long-lived outer-loop agents](https://theouterloop.substack.com/p/openais-realtime-api-is-a-step-towards) that can process asynchronous execution of both LLM inference and long-running tool calls. It's designed for simplicity and gives strong durability and reliability guarantees. It embraces concepts from [12-factor-agents](https://hlyr.dev/12fa) for agents that make asynchronous tool calls like contacting humans or delegating work to other agents. |
| 8 | |
| 9 | :warning: **Note** - ACP is in alpha. |
| 10 | |
| 11 | <div align="center"> |
| 12 | |
| 13 | <h3> |
| 14 | |
| 15 | [Discord](https://humanlayer.dev/discord) | [Documentation](./docs) | [Examples](./acp-example) |
| 16 | |
| 17 | </h3> |
| 18 | |
| 19 | [](https://github.com/humanlayer/agentcontrolplane) |
| 20 | [](https://opensource.org/licenses/Apache-2) |
| 21 | |
| 22 | </div> |
| 23 | |
| 24 | ## Table of Contents |
| 25 | |
| 26 | - [Key Features](#key-features) |
| 27 | - [Architecture](#architecture) |
| 28 | - [Core Objects](#core-objects) |
| 29 | - [Getting Started](#getting-started) |
| 30 | - [Prerequisites](#prerequisites) |
| 31 | - [Setting Up a Local Cluster](#setting-up-a-local-cluster) |
| 32 | - [Deploying ACP](#deploying-acp) |
| 33 | - [Creating an Agent and Running your first task](#creating-an-agent-and-running-your-first-task) |
| 34 | - [Adding Tools with MCP](#adding-tools-with-mcp) |
| 35 | - [Using other language models](#using-other-language-models) |
| 36 | - [Delegating to a Sub-Agent](#delegating-to-a-sub-agent) |
| 37 | - [Incorporating Human Approval](#incorporating-human-approval) |
| 38 | - [Incorporating Humans as Tools](#humans-as-tools) |
| 39 | - [Cleaning Up](#cleaning-up) |
| 40 | - [Design Principles](#design-principles) |
| 41 | - [Contributing](#contributing) |
| 42 | - [License](#license) |
| 43 | |
| 44 | |
| 45 | ## Architecture |
| 46 | |
| 47 | <img width="1918" alt="Screenshot 2025-04-10 at 9 00 58 AM" src="https://github.com/user-attachments/assets/854a2496-0e6c-438f-aee5-6805263fface" /> |
| 48 | |
| 49 | ### Core Objects |
| 50 | |
| 51 | - **LLM**: Provider + API Keys + Parameters |
| 52 | - **Agent**: LLM + System Prompt + Tools |
| 53 | - **Tools**: MCP Servers, Humans, Other Agents |
| 54 | - **Task**: Agent + User Message + Current context window |
| 55 | - **ToolCall**: A single tool call that occurred during a Task |
| 56 | |
| 57 | ## Getting Started |
| 58 | |
| 59 | ### Prerequisites |
| 60 | |
| 61 | To run ACP, you'll need: |
| 62 | |
| 63 | - **kubectl** - Command-line tool for Kubernetes `brew install kubectl` |
| 64 | - **OpenAI API Key** - For LLM functionality https://platform.openai.com |
| 65 | |
| 66 | To run ACP locally on macos, you'll also need: |
| 67 | |
| 68 | - **kind** - For running local Kubernetes clusters `brew install kind` (other cluster options should work too) |
| 69 | - **Docker** - For building and running container images `brew install --cask docker` |
| 70 | |
| 71 | ### Setting Up a Local Cluster |
| 72 | |
| 73 | ```bash |
| 74 | kind create cluster |
| 75 | ``` |
| 76 | |
| 77 | ### Add your OpenAI API key as a Kubernetes secret |
| 78 | |
| 79 | For Anthropic and other providers, see [Using other language models](#using-other-language-models) |
| 80 | |
| 81 | ```bash |
| 82 | kubectl create secret generic openai \ |
| 83 | --from-literal=OPENAI_API_KEY=$OPENAI_API_KEY \ |
| 84 | --namespace=default |
| 85 | ``` |
| 86 | |
| 87 | ### Deploying ACP |
| 88 | |
| 89 | |
| 90 | > [!TIP] |
| 91 | > For better visibility when running tutorial, we recommend starting |
| 92 | > a stream to watch all the events as they're happening, |
| 93 | > for example: |
| 94 | > |
| 95 | > ```bash |
| 96 | > kubectl get events --watch |
| 97 | > ``` |
| 98 | |
| 99 | Deploy the ACP operator to your cluster: |
| 100 | |
| 101 | ```bash |
| 102 | kubectl apply -f https://raw.githubusercontent.com/humanlayer/agentcontrolplane/refs/heads/main/acp/config/release/latest.yaml |
| 103 | ``` |
| 104 | |
| 105 | <details> |
| 106 | <summary>Just the CRDs</summary> |
| 107 | |
| 108 | ```bash |
| 109 | kubectl apply -f https://raw.githubusercontent.com/humanlayer/agentcontrolplane/refs/heads/main/acp/config/release/latest-crd.yaml |
| 110 | ``` |
| 111 | |
| 112 | </details> |
| 113 | |
| 114 | <details> |
| 115 | <summary>Install a specific version</summary> |
| 116 | |
| 117 | ```bash |
| 118 | kubectl apply -f https://raw.githubusercontent.com/humanlayer/agentcontrolplane/refs/heads/main/acp/config/release/v0.1.0.yaml |
| 119 | ``` |
| 120 | |
| 121 | </details> |
| 122 | |
| 123 | This command will build the operator, create necessary CRDs, and deploy |
People who liked this also viewed
6 picksType
- fastmcp🚀 The fast, Pythonic way to build MCP servers and clients.MCP ServersJul 2026
- playwright-mcpPlaywright MCP serverMCP ServersJul 2026
- ext-appsOfficial repo for spec & SDK of MCP Apps protocol - standard for UIs embedded AI chatbots, served by MCP serversMCP ServersJul 2026
- langchain-mcp-adaptersLangChain 🔌 MCPMCP ServersJul 2026
- mcporterCall MCPs via TypeScript, masquerading as simple TypeScript API. Or package them as cli.MCP ServersJul 2026
- mcp-proxyA bridge between Streamable HTTP and stdio MCP transportsMCP ServersJul 2026
