MCP Agent Bridge
Provides web search capability via DuckDuckGo's search engine.
Provides AI image generation using OpenAI's models.
Provides a shared key-value memory store backed by SQLite.
Enables sending messages through Telegram channels.
Enables sending messages through WeChat.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
MCP Agent Bridge
A lightweight MCP (Model Context Protocol) bridge server that enables Hermes Agent and OpenClaw to share tools and memory, forming a cooperative multi-agent system.
Architecture
┌─────────────┐ ┌──────────────────┐ ┌─────────────┐
│ Hermes │ MCP │ Agent Bridge │ CLI/ │ OpenClaw │
│ Agent │◄───────►│ (FastMCP SSE) │◄───────►│ CLI │
│ │ client │ Port :18900 │ HTTP │ │
│ Feishu/WX │ │ │ │ Image/Web │
│ Memory/Plan │ │ Shared Memory │ │ TTS/Video │
└─────────────┘ └──────────────────┘ └─────────────┘Hermes contributes messaging channels (Feishu, WeChat, Telegram…) and its planning/reasoning capabilities.
OpenClaw contributes AI image generation, web search, TTS, and video generation.
The Bridge makes each agent's strengths available to the other through standard MCP tool calls, plus a shared SQLite-backed memory store.
Related MCP server: ACP-MCP-Server
Exposed Tools
Tool | Direction | Description |
| OC → Hermes | AI image generation via OpenClaw |
| OC → Hermes | Web search via OpenClaw |
| OC → Hermes | URL content extraction via OpenClaw |
| OC → Hermes | Text-to-speech via OpenClaw |
| Hermes → OC | Send messages through Hermes's platforms |
| Hermes → OC | Delegate tasks to Hermes's AI agent |
| Bidirectional | Read from shared key-value store |
| Bidirectional | Write to shared key-value store |
| Bidirectional | List memory keys with optional filters |
| Bidirectional | Delete a memory key |
| — | Health check and registered module list |
Quick Start
Prerequisites
Python 3.11+
Hermes Agent with API Server enabled
OpenClaw CLI installed and configured
pip install "mcp[cli]>=1.0" aiohttp pyyaml
Install
git clone https://github.com/fkdt01/mcp-agent-bridge.git
cd mcp-agent-bridge
pip install -e .Configure
cp config.example.yaml config.yaml
# Edit config.yaml — fill in your API keys and pathsKey settings:
openclaw:
cli_path: "openclaw" # or full path like /usr/local/bin/openclaw
image_provider: "openai" # default image gen provider
web_search_provider: "duckduckgo"
hermes:
api_url: "http://127.0.0.1:8888"
api_key: "" # or set HERMES_API_KEY env var
default_channel: "feishu"
memory:
backend: "sqlite"
db_path: "data/bridge_memory.db"Run
# Start the bridge server
python -m bridge.server
# With custom options
python -m bridge.server --config /path/to/config.yaml --port 18900 --log-level DEBUGConnect Hermes
Add to ~/.hermes/config.yaml:
mcp_servers:
agent-bridge:
transport: sse
url: http://127.0.0.1:18900/sseConnect OpenClaw
Add to ~/.openclaw/openclaw.json → mcpServers:
{
"mcpServers": {
"agent-bridge": {
"transport": "sse",
"url": "http://127.0.0.1:18900/sse"
}
}
}Usage Examples
Hermes generates an image via OpenClaw
When Hermes needs to generate an image (e.g. from a Feishu chat), it calls:
oc_image_generate({
prompt: "A futuristic city at sunset, cyberpunk style",
aspect_ratio: "16:9",
model: "openai"
})→ Bridge runs openclaw capability image generate --prompt "..." --json
→ Returns image path/metadata to Hermes
→ Hermes delivers the image to the user
OpenClaw sends a Feishu message via Hermes
hermes_send_message({
message: "✅ Image generation complete!",
target: "feishu"
})→ Bridge POSTs to Hermes API /v1/chat/completions
→ Hermes delivers the message to the Feishu channel
Shared memory between agents
Hermes writes project context:
shared_memory_write({
key: "project.math-game.pet-design",
value: "算术小喵: 草原主题, 绿色配色",
source: "hermes"
})OpenClaw reads it later:
shared_memory_read({ key: "project.math-game.pet-design" })
→ { value: "算术小喵: 草原主题, 绿色配色", source: "hermes", updated_at: 1745631000 }Adding Custom Tools
Create a new file in bridge/tools/ following the naming convention:
oc_*.py— OpenClaw-backed toolshermes_*.py— Hermes-backed toolsshared_*.py— Shared/bidirectional tools
Each module must expose a register(mcp, config) function:
"""My custom tool."""
from mcp.server.fastmcp import FastMCP
from typing import Any
def register(mcp: FastMCP, config: dict[str, Any]) -> None:
@mcp.tool()
async def my_custom_tool(param: str) -> dict[str, Any]:
"""Tool description — this becomes the MCP tool description."""
# Your implementation here
return {"result": "ok"}The tool is auto-discovered and registered on server startup.
Running as a systemd Service
cat > ~/.config/systemd/user/mcp-agent-bridge.service << 'EOF'
[Unit]
Description=MCP Agent Bridge Server
After=network.target
[Service]
Type=simple
WorkingDirectory=/path/to/mcp-agent-bridge
ExecStart=/usr/bin/python -m bridge.server
Restart=on-failure
RestartSec=5
Environment=HERMES_API_KEY=your_key_here
[Install]
WantedBy=default.target
EOF
systemctl --user daemon-reload
systemctl --user enable --now mcp-agent-bridgeProject Structure
mcp-agent-bridge/
├── bridge/
│ ├── __init__.py
│ ├── __main__.py
│ ├── server.py # FastMCP server entry point
│ ├── config.py # YAML + env-var config loader
│ ├── openclaw_runner.py # Shared async CLI subprocess runner
│ ├── memory.py # SQLite-backed shared memory store
│ └── tools/
│ ├── __init__.py # Auto-discovery & registration
│ ├── oc_image.py # Image generation
│ ├── oc_web.py # Web search & fetch
│ ├── oc_tts.py # Text-to-speech
│ ├── hermes_messaging.py # Message sending
│ ├── hermes_chat.py # Task delegation
│ └── shared_memory.py # Shared memory tools
├── config.example.yaml
├── .gitignore
├── LICENSE
├── README.md
└── pyproject.tomlLicense
MIT
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
The Remote MCP server acts as a standardized bridge between LLM applications (like Claude, ChatGPT, and Cursor) and external services, enabling AI agents to access external tools and resources. Its primary capability is providing a centralized search tool to discover other MCP servers and their respective tools. Unlike local implementations, it runs remotely with OAuth authentication and permission controls for security.
An MCP memory server. One memory your agents share — across models, devices and apps.
Infrastructure for AI agents from different companies and people to work together.
Shared memory and mail for your AI agents. Verified with Claude Code; other MCP clients in testing.
Related MCP Servers
- AlicenseAqualityFmaintenanceA bridge server that enables MCP-compatible AI assistants like Claude to seamlessly discover, communicate with, and manage A2A protocol agents.7113 PyPI146Apache 2.0
- AlicenseBqualityFmaintenanceA bridge server that connects Agent Communication Protocol (ACP) agents with Model Context Protocol (MCP) clients, enabling seamless integration between ACP-based AI agents and MCP-compatible tools like Claude Desktop.1665 PyPI24MIT
- AlicenseNot gradedqualityDmaintenanceAn MCP server and local HTTP bridge designed to integrate remote upstream MCP tools into OpenClaw skills or local environments. It enables users to generate skill wrappers and proxy tool calls via a local HTTP bridge for use in Claude Desktop, Cursor, or OpenClaw.Apache 2.0
- AlicenseBqualityNot gradedmaintenanceBridges OpenClaw and Hermes Agent, enabling multi-turn conversations, messaging via Hermes channels, and health checks through MCP tools.4MIT