Search public document metadata and pinned snapshot provenance, excluding passages and identifiers.
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.
Read one public source document ID, source-reported change counts, classifications and snapshot provenance. No names, passages, private identifiers or invented verification.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
claude mcp add --transport http unredactus https://unredactus.com/api/mcp
Read one public source document ID, source-reported change counts, classifications and snapshot provenance. No names, passages, private identifiers or invented verification.
| Argument | Type | Description |
|---|---|---|
| documentId* | string |
{
"type": "object",
"required": [
"documentId"
],
"properties": {
"documentId": {
"type": "string",
"pattern": "^EFTA[0-9]{8}$",
"maxLength": 12
}
},
"additionalProperties": false
}Read snapshot dates, pinned upstream commit, source hashes, license, counts and evidence limitations. Not a live monitor or independent verification.
No arguments.
{
"type": "object",
"required": [],
"properties": {},
"additionalProperties": false
}Find bounded public document metadata by document-ID prefix, dataset or source classification. No person search or removed-text passages. Preserve automated/triage labels and false-positive states.
| Argument | Type | Description |
|---|---|---|
| limit | integer | |
| offset | integer | |
| dataset | integer | |
| category | string | |
| documentPrefix | string |
{
"type": "object",
"required": [],
"properties": {
"limit": {
"type": "integer",
"maximum": 20,
"minimum": 1
},
"offset": {
"type": "integer",
"maximum": 30000
},
"dataset": {
"enum": [
1,
3,
8,
9,
10,
11,
12
],
"type": "integer",
"maximum": 12,
"minimum": 1
},
"category": {
"enum": [
"CONTENT_REMOVAL",
"OCR_IMPROVEMENT",
"MIXED",
"METADATA_CHANGE",
"NAME_REDACTION",
"EMAIL_REDACTION",
"ADDRESS_REDACTION",
"DOB_REDACTION",
"PHONE_REDACTION",
"ACCOUNT_REDACTION",
"UNCLASSIFIED",
"NO_CHANGE",
"NON_PII",
"SSN_REDACTION",
"PRIVACY_ACT_PII",
"CONTENT_REMOENTION",
"PII_REDACTION"
],
"type": "string"
},
"documentPrefix": {
"type": "string",
"pattern": "^EFTA[0-9]{0,8}$",
"maxLength": 12
}
},
"additionalProperties": false
}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 Oct 31, 2026.