- Home
- MCP servers
- DeepRecall
DeepRecall
- python
0
GitHub Stars
python
Language
4 months ago
First Indexed
2 months ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
You can access DeepRecall’s MCP server to perform rapid product safety recalls searches from multiple regulatory sources, integrate risk checks into agent workflows, and make informed checkout decisions with confidence. This MCP server connects your client to recalls data and offers practical tools to search recalls by text or images and to list available data sources.
How to use
This MCP server provides two core ways to connect from your client. First, you can use the hosted server by adding a remote MCP URL to your client configuration. Second, you can run the server locally using Python if you prefer self-hosting. Once connected, you can search recalls using descriptive queries or images and you can explore which data sources are available for recalls.
How to install
Prerequisites: you need Python 3.8+ and Git installed on your system. You may also use Pip to install Python dependencies.
Option A: Use the hosted MCP server (recommended) without running your own server. No local installation is required beyond your MCP client configuration.
Option B: Self-host the MCP server locally.
Step-by-step to self-host:
git clone https://github.com/adrida/deeprecall-mcp
cd deeprecall-mcp
pip install -r requirements.txt
DEEPRECALL_API_KEY=your_key python server.py --http
Configuration and usage notes
Configure the hosted MCP server in your client with the following endpoint. Replace the API key with your own key.
{
"mcpServers": {
"deeprecall": {
"url": "https://mcp.deeprecall.io/mcp",
"headers": {
"X-API-Key": "dr_live_your_api_key_here"
}
}
}
}
Data sources and tools
The server exposes tools to search recalls and to list data sources. You can use them to retrieve recall information by description or by image, and to see which regulatory agencies are included.
Docker
If you prefer containerized deployment, you can build and run a container image with the standard runtime configuration.
docker build -t deeprecall-mcp .
docker run -p 8000:8000 -e DEEPRECALL_API_KEY=your_key deeprecall-mcp
UCP Integration
This MCP server is designed for Universal Commerce Protocol (UCP) integration, providing the product safety layer for agentic commerce. The flow connects Agent Intent to Product Discovery, then to DeepRecall Safety Check, and finally to Checkout Decision.
See the UCP capability definition for detailed requirements and how to wire in this safety layer.
API access and notes
API access is currently in closed beta. Access is available to beta testers by invitation and to enterprise clients. Contact the team to request access.
Prices and plan options are announced for beta, business, and enterprise tiers, with varying levels of usage and support.
Available tools
search_recalls
Search for recalled products by text descriptions or image URLs, returning matching recalls from multiple agencies.
get_data_sources
List available regulatory agencies contributing recall data to the MCP server.