Metadata-Version: 2.4
Name: aimarket-mcp-packager
Version: 2.1.2
Summary: Package products as ready-to-run MCP servers for Claude Desktop
License: MIT
Requires-Python: >=3.11
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: aimarket-hub>=3.0.0
Requires-Dist: mcp<2,>=1.6
Provides-Extra: mcp
Provides-Extra: dev
Requires-Dist: pytest>=8; extra == "dev"
Requires-Dist: jsonschema>=4.23; extra == "dev"
Dynamic: license-file

# aimarket-mcp-packager — MCP server

<!-- mcp-name: io.github.alexar76/aimarket-plugins -->

<!-- aicom-readme-badges -->
<p align="center">
  <a href="https://github.com/alexar76/aimarket-plugins/actions/workflows/ci.yml"><img src="https://github.com/alexar76/aimarket-plugins/actions/workflows/ci.yml/badge.svg" alt="CI" /></a>
  <a href="https://glama.ai/mcp/servers/alexar76/aimarket-plugins"><img src="https://glama.ai/mcp/servers/alexar76/aimarket-plugins/badges/score.svg" alt="aimarket-plugins MCP server" /></a>
  <a href="docs/badges/coverage.svg"><img src="docs/badges/coverage.svg" alt="Test coverage" /></a>
  <a href="LICENSE"><img src="docs/badges/license.svg" alt="License: Apache-2.0" /></a>
</p>
<!-- /aicom-readme-badges -->

Quality score: **[glama.ai/mcp/servers/alexar76/aimarket-plugins/score](https://glama.ai/mcp/servers/alexar76/aimarket-plugins/score)** ·
[![alexar76/aimarket-plugins MCP server](https://glama.ai/mcp/servers/alexar76/aimarket-plugins/badges/score.svg)](https://glama.ai/mcp/servers/alexar76/aimarket-plugins)

**This repository ships a [Model Context Protocol (MCP)](https://modelcontextprotocol.io) server** that packages AIMarket capabilities into self-hosted MCP server products (Docker image metadata, MCP manifest, Claude Desktop config).

Transport: **stdio** (`aimarket_mcp_packager/stdio_server.py`). Built with the official **[MCP Python SDK](https://github.com/modelcontextprotocol/python-sdk)** (`mcp` package, `FastMCP`).

Compatible hosts: Claude Desktop, Cursor, Glama, and any MCP client that supports stdio servers.

---

## MCP server in this repo

| Item | Location |
|------|----------|
| MCP entrypoint | `aimarket_mcp_packager/stdio_server.py` |
| MCP SDK dependency | Installed by default — `mcp>=1.6,<2` |
| Glama / Docker run | [`Dockerfile`](Dockerfile), [`glama.json`](glama.json), [docs/GLAMA.md](docs/GLAMA.md) |
| Tool implementation | [`aimarket_mcp_packager/packager_core.py`](aimarket_mcp_packager/packager_core.py) |

Run locally:

```bash
pip install -r requirements-mcp.txt
pip install -e .
python -m aimarket_mcp_packager.stdio_server
```

Or with Docker (stdio MCP server):

```bash
docker build -t aimarket-mcp-packager .
docker run --rm -i aimarket-mcp-packager
```

---

## MCP tools

The server exposes **3 tools** (no MCP resources or prompts):

| Tool | Description |
|------|-------------|
| `package_capability` | Build an MCP server package: Docker image name, `mcp_manifest`, subscription tiers, connection string |
| `generate_dockerfile` | Generate a self-contained Dockerfile with a working stdio capability relay |
| `generate_claude_desktop_config` | Generate a `claude_desktop_config.json` snippet (`mcpServers`) for docker-run MCP |

Example tool call (conceptual):

```json
{
  "name": "package_capability",
  "arguments": {
    "capability_id": "translate.multi@v2",
    "product_id": "prod-001",
    "name": "Lyra",
    "description": "Multilingual translator",
    "input_schema": { "type": "object", "properties": { "text": { "type": "string" } }, "required": ["text"] }
  }
}
```

> `input_schema` takes a JSON Schema **object** (a JSON string is also tolerated for backward compatibility). Every tool parameter is documented in the server's tool schema for MCP clients.

The generated Dockerfile includes the server source and installs its runtime; build it
with the image tag returned by `package_capability`. Both client-config generators use
attached `docker run --rm -i` so MCP messages travel over stdio. Set `AIMARKET_HUB_URL`
and, for capabilities on another provider, `AIMARKET_SOURCE_HUB`. An optional
`AIMARKET_API_KEY` pays calls through the Hub; these environment variables are forwarded
at runtime and are never embedded in the image. The example subscription tiers are
metadata only: this relay does not implement subscription billing or license checks.

---

## MCP resources

This server does **not** register MCP resources or prompts — only the tools above. Output is JSON text returned from each tool call.

---

## Client configuration

### Claude Desktop

```json
{
  "mcpServers": {
    "aimarket-mcp-packager": {
      "command": "python",
      "args": ["-m", "aimarket_mcp_packager.stdio_server"]
    }
  }
}
```

Compat path (what Glama's admin form historically used — keep the shim file):

```json
{
  "mcpServers": {
    "aimarket-mcp-packager": {
      "command": "python",
      "args": ["plugins/aimarket-mcp-packager/mcp_stdio_server.py"]
    }
  }
}
```

Docker variant (see `generate_claude_desktop_config` tool output):

```json
{
  "mcpServers": {
    "aimarket-mcp-packager": {
      "command": "docker",
      "args": ["run", "--rm", "-i", "aimarket-mcp-packager:latest"]
    }
  }
}
```

### Cursor

Add the same stdio server under **Settings → MCP** (command + args pointing at `aimarket_mcp_packager/stdio_server.py` or the Docker image above).

---

## Python library (optional)

The same packaging logic is available as a library and as an AIMarket Hub plugin (PyPI package coming soon — for now `pip install -e .` from this directory, or use the AIMarket Hub Docker image). Hub HTTP routes are optional; **the MCP server runs standalone** via `aimarket_mcp_packager/stdio_server.py`.

```python
from aimarket_mcp_packager.mcp_packager import MCPPackager

packager = MCPPackager()
pkg = packager.package(
    capability_id="translate.multi@v2",
    product_id="prod-001",
    name="Lyra",
    description="Multilingual translator",
    input_schema={"type": "object", "properties": {"text": {"type": "string"}}},
    price_per_call_usd=0.40,
)

print(pkg.docker_image)
print(pkg.mcp_manifest)
print(packager.generate_claude_desktop_config(pkg))
```

---

## When to use

- Distribute AI capabilities as self-hosted MCP servers (Docker + manifest)
- Generate Claude Desktop / Cursor MCP client configs
- Enterprise on-prem deployment with subscription tiers

## Subscription tiers (packaged products)

| Tier | Calls/month | Price |
|------|------------|-------|
| Starter | 100 | $9.99 |
| Pro | 1,000 | $49.99 |
| Enterprise | 10,000 | $299.99 |

## Documentation

| Document | Description |
|----------|-------------|
| [User guide](docs/user-guide.md) | Install, Hub plugin, verify loaded |
| [User cases](docs/user-cases.md) | Personas and cross-plugin workflows |
| [SDK integration](docs/sdk-integration.md) | Code examples and hook behavior |

## License

MIT · Maintained by AI-Factory
