Look up the USDA plant behind any meat or poultry package code: FSIS Salmonella categories, recalls.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Full public-data profile for one establishment: registry identity, Salmonella verification history, lab sampling results, recalls, and a neutral reported-status summary with the USDA source URLs behind it.
This tool isn't marked read-only, so it may change something on the server or in an account you connected.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http usda-establishments https://foodsafetydata-api.onrender.com/mcp
Full public-data profile for one establishment: registry identity, Salmonella verification history, lab sampling results, recalls, and a neutral reported-status summary with the USDA source URLs behind it.
| Argument | Type | Description |
|---|---|---|
| code* | string |
{
"type": "object",
"title": "get_safety_profileArguments",
"required": [
"code"
],
"properties": {
"code": {
"type": "string",
"title": "Code"
}
}
}Resolve a USDA inspection code from a package label (e.g. P12345, M12345, I135 or a company name) to its establishment: plant name, company, location, status, Salmonella verification categories per product class, and recall history.
| Argument | Type | Description |
|---|---|---|
| code* | string |
{
"type": "object",
"title": "lookup_establishmentArguments",
"required": [
"code"
],
"properties": {
"code": {
"type": "string",
"title": "Code"
}
}
}Search the USDA establishment registry by inspection code, company or plant name, or city — returns matching establishments with codes and profile URLs.
| Argument | Type | Description |
|---|---|---|
| q* | string | |
| limit | integer |
{
"type": "object",
"title": "search_establishmentsArguments",
"required": [
"q"
],
"properties": {
"q": {
"type": "string",
"title": "Q"
},
"limit": {
"type": "integer",
"title": "Limit",
"default": 8
}
}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 19, 2026.
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.