home_mcp_gateway
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@home_mcp_gatewayGenerate an image of a futuristic city"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
home_mcp_gateway
home_mcp_gateway is a local HTTP/SSE MCP gateway for ZeroClaw and other MCP clients. It centralizes tool calls, policy checks, artifact storage, jobs, and audit records behind one local MCP endpoint.
Manage settings, agents and LLM providers/models in the WebUI. Management data is stored in SQLite (default: artifacts/metadata.sqlite3). The first startup imports existing YAML and remote image settings; YAML remains the bootstrap configuration source.
Quick Start
Requirements:
Docker with Docker Compose
PowerShell on Windows for the helper scripts
Edit config/config.yaml and set at least callers.host_assistant.token, callers.role_default.token, and artifacts.signed_url_secret.
Build and run with Docker Compose:
$env:HOME_MCP_LOGIN_KEY = "your-login-key"
docker compose up -d --buildThe default endpoint is http://127.0.0.1:8787.
Log in to the WebUI with HOME_MCP_LOGIN_KEY, independent of MCP caller tokens. Compose also accepts this variable from .env. In LLM Management, configure an OpenAI-compatible provider URL and API key, fetch models, select models to add, and assign text / image tags. URLs with /v1 are accepted. Remote Image selects an added model tagged image.
Check health:
Invoke-RestMethod http://127.0.0.1:8787/healthz
Invoke-RestMethod http://127.0.0.1:8787/readyzRelated MCP server: jgkme/kilo-image-gen-mcp
MCP Client
Use the SSE transport URL:
[[mcp.servers]]
name = "home"
transport = "sse"
url = "http://127.0.0.1:8787/mcp"
deferred_loading = trueWhen another Docker Compose service connects to this gateway on the same Docker network, use:
[[mcp.servers]]
name = "home"
transport = "sse"
url = "http://home-mcp:8787/mcp"
deferred_loading = trueLocal Python Development
python -m pip install -e .
$env:HOME_MCP_DEV_MODE = "true"
python -m app.mainHOME_MCP_DEV_MODE=true opens management without login. MCP caller authentication and policy still apply. Use HOME_MCP_LOGIN_KEY with dev mode off for normal operation. Quotas, windows, concurrency, policy, artifacts, agents and tool parameters apply immediately on save. Only module switches require a restart.
The Settings page manages image/TTS/print quotas, counting windows, concurrency, artifact expiry and MCP access policy. The default image window is 86400 seconds (24 hours); both quota and window are editable. Zero quota/concurrency values mean unlimited. Module pages expose Parameters & limits for image editing/downloads, ComfyUI presets, voice options and print defaults. Running jobs keep their original settings, and in-memory quota counters reset on restart.
The default path is Docker Compose. Use local Python runs for development only.
Agents
Manage agents in the WebUI; caller and Matrix tokens are saved to SQLite. After migration, YAML edits do not override saved management data. Back up the database together with artifacts.
Tests
.\tests\run_tests.ps1Documentation
Documentation index: docs/README.md
User guide: docs/user/README.md
Current developer docs: docs/developer/README.md
Original development docs: docs/original/README.md
Chinese quick start: README.zh-cn.md
This server cannot be deployed
Maintenance
Related MCP Connectors
Zero-setup MCP gateway securely connecting AI to your tools with authentication and workflows
- gatewayOAuthai.sealgate
MCP gateway with runtime security policy, tool-call-level control, and audit of agent actions.
MCP server for Qwen Image 3 AI image generation
Generate images from Claude, ChatGPT, Cursor and Codex with 12 top AI image models. Up to 4 images per call, SVG vectorization, background removal, upscaling and image-to-video. Remote MCP with OAuth, no API keys.
Related MCP Servers
- AlicenseAqualityDmaintenanceMCP server for multi-provider AI image generation (AWS Bedrock, OpenAI, Google Gemini) enabling image generation, transformation, and editing through a unified interface.41MIT
- AlicenseNot gradedqualityCmaintenanceMCP server for generating, editing, and processing images via multiple providers including Kilo, OpenRouter, OpenAI, and Gemini, with local tools for background removal, resizing, and cropping.10 npm3MIT
- AlicenseAqualityDmaintenanceMCP server that provides image generation, captioning, and tagging via ComfyUI API, configurable for agent tools.43MIT
- AlicenseAqualityCmaintenanceMCP server that connects Codex or MCP clients to MetaRouter's OpenAI-compatible image generation API, enabling natural language image generation, local image editing, batch editing, and multi-reference image synthesis.5MIT