
Dated, source-linked public records about companies, from SEC filings and other primary documents.
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.
Get one covered company by name, former name, SEC CIK or slug: identity (legal name, former names, state of incorporation, SEC EDGAR source), dossier status, a plain-language coverage statement saying which record sections are on file and which are NOT MONITORED (not monitored is not zero), all of its records (reviewed and unreviewed, each with tier), and up to 5 recent SEC filings from the wire. A name matching several companies returns an error listing them; a company outside coverage returns an error (isError) that says so.
Company name, former name, SEC CIK (leading zeros optional) or slug, e.g. "Cipher Digital" or "1819989".
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http public-records https://seraphixsystems.com/mcp
Get one covered company by name, former name, SEC CIK or slug: identity (legal name, former names, state of incorporation, SEC EDGAR source), dossier status, a plain-language coverage statement saying which record sections are on file and which are NOT MONITORED (not monitored is not zero), all of its records (reviewed and unreviewed, each with tier), and up to 5 recent SEC filings from the wire. A name matching several companies returns an error listing them; a company outside coverage returns an error (isError) that says so.
| Argument | Type | Description |
|---|---|---|
| name_or_id* | string | Company name, former name, SEC CIK (leading zeros optional) or slug, e.g. "Cipher Digital" or "1819989". |
{
"type": "object",
"required": [
"name_or_id"
],
"properties": {
"name_or_id": {
"type": "string",
"minLength": 1,
"description": "Company name, former name, SEC CIK (leading zeros optional) or slug, e.g. \"Cipher Digital\" or \"1819989\"."
}
},
"additionalProperties": false
}Describe exactly what Seraphix covers and what it does not, from the published data: companies tracked and their dossier status, states, record counts by tier, each source with its status (OBSERVED, DEGRADED or NOT MONITORED) and last good check, an explicit not-covered list, and the publication rules (source links, company-stated labels, corrections). Call this before concluding that something is absent. Takes no arguments.
No arguments.
{
"type": "object",
"properties": {},
"additionalProperties": false
}List SEC EDGAR filings accepted for the companies Seraphix covers, newest first: company, CIK, form, 8-K item codes and titles, SEC acceptance time (source_date), accession number and the SEC filing URL (source_url). Insider and ownership forms (3, 4, 5, 144, SC 13D/G) are excluded. Notices are machine-extracted index metadata (tier "machine-extracted"), not reviewed records; read the filing itself before relying on it. Optional filters: company, since (acceptance time on or after an ISO 8601 date or date-time), limit. If the wire is unavailable the tool returns an error (isError) with its last good check instead of an empty list.
| Argument | Type | Description |
|---|---|---|
| limit | integer | Maximum notices (1-100, default 25). |
| since | string | Only filings accepted on or after this ISO 8601 date or date-time, e.g. 2026-11-01 or 2026-11-05T13:00:00Z. |
| company | string | Restrict to one covered company: name, SEC CIK or slug. |
{
"type": "object",
"properties": {
"limit": {
"type": "integer",
"default": 25,
"maximum": 100,
"minimum": 1,
"description": "Maximum notices (1-100, default 25)."
},
"since": {
"type": "string",
"description": "Only filings accepted on or after this ISO 8601 date or date-time, e.g. 2026-11-01 or 2026-11-05T13:00:00Z."
},
"company": {
"type": "string",
"description": "Restrict to one covered company: name, SEC CIK or slug."
}
},
"additionalProperties": false
}Fetch one Seraphix public record by ID, with everything needed to cite and check it: the claim, the verbatim quote, source_url, source title, publisher and locator, source_date, published_at, the primary document, supporting quotes from other filings, the dated history of filings stating the same thing, every version (superseded versions keep their reason and correction source), tier, and a ready-made citation. Record IDs look like SX-07-A-0001. An unknown ID returns an error (isError) saying so; IDs are never reused.
| Argument | Type | Description |
|---|---|---|
| id* | string | Record ID, e.g. SX-07-A-0001. |
{
"type": "object",
"required": [
"id"
],
"properties": {
"id": {
"type": "string",
"pattern": "^[Ss][Xx]-[A-Za-z0-9-]+$",
"description": "Record ID, e.g. SX-07-A-0001."
}
},
"additionalProperties": false
}List every company in the Seraphix dossier index (the only companies Seraphix covers), alphabetically, with SEC CIK, role, dossier status (FULL = reviewed sections on file; STUB = identity and SEC filing metadata only), the number of reviewed records, the latest record date, the company page URL and its SEC EDGAR source URL. Takes no arguments.
No arguments.
{
"type": "object",
"properties": {},
"additionalProperties": false
}Search Seraphix public records: dated, source-linked statements taken from primary documents (SEC filings, statutes, official codes) about the companies Seraphix covers and state data-center policy. Matching is case-insensitive; every word of `query` must appear in the record (claim, verbatim quote, title, company, source title or fields). Results are newest record_date first. By default only reviewed records are returned; pass tier "all" to include unreviewed ones. Each result has the record URL to cite, source_url, source_date, published_at, tier and version. An empty result means no matching record in Seraphix coverage, not that no such fact exists; call get_coverage to see what is covered. No news, prices or information about individuals.
| Argument | Type | Description |
|---|---|---|
| tier | string | "reviewed" (default) or "all" to include unreviewed records. |
| type | string | Record section: sites, power, incentives, water_permits or policy. |
| limit | integer | Maximum results (1-50, default 10). |
| query | string | Words to match, e.g. "Barber Lake" or "300 MW". Omit to list records by the other filters. |
| since | string | Only records whose record_date is on or after this ISO 8601 date, e.g. 2026-01-01. |
| company | string | Restrict to one covered company: name, former name, SEC CIK or slug (see list_companies). |
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 6, 2026.
{
"type": "object",
"properties": {
"tier": {
"enum": [
"reviewed",
"all"
],
"type": "string",
"default": "reviewed",
"description": "\"reviewed\" (default) or \"all\" to include unreviewed records."
},
"type": {
"enum": [
"sites",
"power",
"incentives",
"water_permits",
"policy"
],
"type": "string",
"description": "Record section: sites, power, incentives, water_permits or policy."
},
"limit": {
"type": "integer",
"default": 10,
"maximum": 50,
"minimum": 1,
"description": "Maximum results (1-50, default 10)."
},
"query": {
"type": "string",
"maxLength": 200,
"description": "Words to match, e.g. \"Barber Lake\" or \"300 MW\". Omit to list records by the other filters."
},
"since": {
"type": "string",
"description": "Only records whose record_date is on or after this ISO 8601 date, e.g. 2026-01-01."
},
"company": {
"type": "string",
"description": "Restrict to one covered company: name, former name, SEC CIK or slug (see list_companies)."
}
},
"additionalProperties": false
}