Cross-brand fault code lookup over one normalized, cited registry.
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.
List the brands in the registry with code counts and the vertical surface that owns each brand.
This tool takes no arguments.
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 registry https://registry.kasafox.com/mcp
List the brands in the registry with code counts and the vertical surface that owns each brand.
No arguments.
{
"type": "object",
"properties": {}
}Look up one fault/error/status code across all KasaFox registry brands and return the normalised meaning, causes, checks, fix steps, parts, safety flags, and citations. Cite the source and hard-stop on safety-critical faults (requires_qualified_installer).
| Argument | Type | Description |
|---|---|---|
| code* | string | The code shown on the display or app, e.g. 'A20', 'Err 1', 'E10379' |
| brand* | string | Brand name |
| model | string | Optional model or series |
| symptom | string | Optional free-text symptom |
{
"type": "object",
"required": [
"brand",
"code"
],
"properties": {
"code": {
"type": "string",
"description": "The code shown on the display or app, e.g. 'A20', 'Err 1', 'E10379'"
},
"brand": {
"enum": [
"barudan",
"domino",
"fronius",
"growatt",
"sma",
"solaredge",
"swf",
"tajima",
"victron",
"videojet"
],
"type": "string",
"description": "Brand name"
},
"model": {
"type": "string",
"description": "Optional model or series"
},
"symptom": {
"type": "string",
"description": "Optional free-text symptom"
}
}
}Search the registry by code, brand, or title/meaning text and return matching rows with their canonical URLs.
| Argument | Type | Description |
|---|---|---|
| q* | string | Search query: a code (e.g. 'a20'), a brand, or a symptom phrase |
{
"type": "object",
"required": [
"q"
],
"properties": {
"q": {
"type": "string",
"description": "Search query: a code (e.g. 'a20'), a brand, or a symptom phrase"
}
}
}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 22, 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.