# Agent Pattern MCP server

> Provides AI agent system design expertise by analyzing requirements to select patterns, generate architectures, and evaluate quality attributes.

- Repository: https://github.com/olk/agent-pattern-mcp
- Page: https://hostedmcp.org/servers/olk/agent-pattern-mcp
- Transport: stdio
- Stars: 0
- License: MIT
- Latest version: 1.0.4

## Install

### Claude Code

```sh
claude mcp add --env 'GENERATOR_API_KEY=<YOUR_GENERATOR_API_KEY>' --env 'EMBEDDER_API_KEY=<YOUR_EMBEDDER_API_KEY>' agent-pattern-mcp -- docker run -i --rm docker.io/olkowa/agent-pattern-mcp:1.0.4
```

### Cursor (`~/.cursor/mcp.json`)

```json
{
  "mcpServers": {
    "agent-pattern-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "docker.io/olkowa/agent-pattern-mcp:1.0.4"
      ],
      "env": {
        "GENERATOR_API_KEY": "<YOUR_GENERATOR_API_KEY>",
        "EMBEDDER_API_KEY": "<YOUR_EMBEDDER_API_KEY>"
      }
    }
  }
}
```

### VS Code (`.vscode/mcp.json`)

```json
{
  "servers": {
    "agent-pattern-mcp": {
      "type": "stdio",
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "docker.io/olkowa/agent-pattern-mcp:1.0.4"
      ],
      "env": {
        "GENERATOR_API_KEY": "<YOUR_GENERATOR_API_KEY>",
        "EMBEDDER_API_KEY": "<YOUR_EMBEDDER_API_KEY>"
      }
    }
  }
}
```

### VS Code CLI

```sh
code --add-mcp '{"name":"agent-pattern-mcp","type":"stdio","command":"docker","args":["run","-i","--rm","docker.io/olkowa/agent-pattern-mcp:1.0.4"],"env":{"GENERATOR_API_KEY":"<YOUR_GENERATOR_API_KEY>","EMBEDDER_API_KEY":"<YOUR_EMBEDDER_API_KEY>"}}'
```

### Claude Desktop (`claude_desktop_config.json`)

```json
{
  "mcpServers": {
    "agent-pattern-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "docker.io/olkowa/agent-pattern-mcp:1.0.4"
      ],
      "env": {
        "GENERATOR_API_KEY": "<YOUR_GENERATOR_API_KEY>",
        "EMBEDDER_API_KEY": "<YOUR_EMBEDDER_API_KEY>"
      }
    }
  }
}
```

### Codex

```sh
codex mcp add agent-pattern-mcp --env 'GENERATOR_API_KEY=<YOUR_GENERATOR_API_KEY>' --env 'EMBEDDER_API_KEY=<YOUR_EMBEDDER_API_KEY>' -- docker run -i --rm docker.io/olkowa/agent-pattern-mcp:1.0.4
```

### Codex config.toml (`~/.codex/config.toml`)

```toml
[mcp_servers.agent-pattern-mcp]
command = "docker"
args = ["run", "-i", "--rm", "docker.io/olkowa/agent-pattern-mcp:1.0.4"]

[mcp_servers.agent-pattern-mcp.env]
GENERATOR_API_KEY = "<YOUR_GENERATOR_API_KEY>"
EMBEDDER_API_KEY = "<YOUR_EMBEDDER_API_KEY>"
```

### Gemini CLI

```sh
gemini mcp add -s user -e 'GENERATOR_API_KEY=<YOUR_GENERATOR_API_KEY>' -e 'EMBEDDER_API_KEY=<YOUR_EMBEDDER_API_KEY>' agent-pattern-mcp docker -- run -i --rm docker.io/olkowa/agent-pattern-mcp:1.0.4
```

### Gemini settings.json (`~/.gemini/settings.json`)

```json
{
  "mcpServers": {
    "agent-pattern-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "docker.io/olkowa/agent-pattern-mcp:1.0.4"
      ],
      "env": {
        "GENERATOR_API_KEY": "<YOUR_GENERATOR_API_KEY>",
        "EMBEDDER_API_KEY": "<YOUR_EMBEDDER_API_KEY>"
      }
    }
  }
}
```

## Environment variables

| Name | Required | Secret | Description |
|---|---|---|---|
| `GENERATOR_API_KEY` | yes | yes | API key for the generator LLM (LiteLLM provider). |
| `EMBEDDER_API_KEY` | no | yes | API key for the embedder. Required for startup even with keyless TEI; use 'not-needed' there. |
| `EMBEDDER_BASE_URL` | no | no | Embedder endpoint. For TEI sidecars on the host use http://host.docker.internal:8080/v1 (add --add-host=host.docker.internal:host-gateway on Linux). |
| `EMBEDDER_PROVIDER` | no | no | Embedding backend for pattern retrieval: tei, openai, ollama, vllm or hosted_vllm. |
| `GENERATOR_BASE_URL` | no | no | OpenAI-compatible base URL (LiteLLM api_base). |
| `GENERATOR_MODEL` | no | no | Model name, combined as <provider>/<model> via LiteLLM. |
| `GENERATOR_PROVIDER` | no | no | LiteLLM provider prefix, e.g. openai, anthropic, openrouter, minimax. |
| `MINIMAXAI_API_KEY` | no | no |  |
| `PATTERN_DIRECTORY` | no | no | Directory with *-pattern.json files. The image bakes the 61-pattern catalog at /app/pattern. |
| `REASONING_ENABLED` | no | no |  |
| `REASONING_FAIL_FAST` | no | no |  |
| `RERANKER_BASE_URL` | no | no | TEI cross-encoder reranker endpoint used during pattern retrieval. |

## Tools (9)

- `analyze_agent_system`
- `generate_agent_system`
- `evaluate_agent_system`
- `design_agent_system`
- `submit_agent_design_job`
- `get_agent_design_status`
- `cancel_agent_design`
- `list_agent_patterns`
- `get_agent_pattern`

## FAQ

### What can the server do for coding agents?

Coding agents can call analyze_agent_system to receive topology and scored-pattern guidance without having to generate a full design themselves.

### How many agent patterns are available?

The server includes 68 built-in agent patterns.

### How do I handle long-running design tasks in clients with short timeouts?

Use the async job pattern by calling submit_agent_design_job to get a job_id, then poll get_agent_design_status until the status is completed.

### What quality attributes does the server evaluate?

The server evaluates designs against reliability, cost efficiency, latency, output quality, observability, safety, and simplicity.
