Legal answers from official sources; refuses when no norm is verified.
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.
Describe a legal situation; get the verified provisions that apply with official-source links. jurisdiction: ru|in|us|uk|ca|hk. Returns a refusal when the corpus has no norm.
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 ailegal-mcp https://ailegal.legal/api/mcp/
Describe a legal situation; get the verified provisions that apply with official-source links. jurisdiction: ru|in|us|uk|ca|hk. Returns a refusal when the corpus has no norm.
| Argument | Type | Description |
|---|---|---|
| text* | string | |
| locale | string | |
| jurisdiction | string |
{
"type": "object",
"title": "legal_breakdownArguments",
"required": [
"text"
],
"properties": {
"text": {
"type": "string",
"title": "Text"
},
"locale": {
"type": "string",
"title": "Locale",
"default": "en"
},
"jurisdiction": {
"type": "string",
"title": "Jurisdiction",
"default": "ru"
}
}
}Legal Oracle: which verified provisions relate to a query. Optional jurisdiction filter. An empty list means the corpus has no norm for this query.
| Argument | Type | Description |
|---|---|---|
| q* | string | |
| limit | integer | |
| jurisdiction | string |
{
"type": "object",
"title": "oracle_lookupArguments",
"required": [
"q"
],
"properties": {
"q": {
"type": "string",
"title": "Q"
},
"limit": {
"type": "integer",
"title": "Limit",
"default": 5
},
"jurisdiction": {
"type": "string",
"title": "Jurisdiction",
"default": ""
}
}
}Legal Oracle: does this citation exist in the verified corpus? Use before relying on any code/article pair. found=false means do not rely on it.
| Argument | Type | Description |
|---|---|---|
| code* | string | |
| article | string | |
| jurisdiction | string |
{
"type": "object",
"title": "verify_citationArguments",
"required": [
"code"
],
"properties": {
"code": {
"type": "string",
"title": "Code"
},
"article": {
"type": "string",
"title": "Article",
"default": ""
},
"jurisdiction": {
"type": "string",
"title": "Jurisdiction",
"default": ""
}
}
}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 5, 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.