EAD Factory MCP Server

MCP server for EAD Factory: qualified evidence, signature, notifications and chat via AI agents.

Quick start

npx -y @g-digital/mcp-ead-factory

Or see ONBOARDING.md for a step-by-step setup guide (≤ 5 minutes).

Where to install

This MCP is published to every major MCP distribution channel by the g-digital MCP distribution pipeline. Pick whichever fits your stack:

Channel	Install command / URL
npm	npx -y @g-digital/mcp-ead-factory — npmjs.com/package/@g-digital/mcp-ead-factory
Docker Hub	docker pull gdigital/ead-factory:latest — hub.docker.com/r/gdigital/ead-factory
MCP Official Registry	Auto-discovered as io.github.g-digital-by-Garrigues/ead-factory — registry.modelcontextprotocol.io
n8n community node	Install @g-digital/n8n-nodes-ead-factory in n8n Settings → Community Nodes — npmjs.com/package/@g-digital/n8n-nodes-ead-factory
Smithery	smithery mcp install g-digital/ead-factory — smithery.ai/servers/g-digital/ead-factory

Every channel ships the same MCP server contract; the tools and environment configuration below apply regardless of which install path you choose.

Need credentials? Visit: https://digitaltrust.gcloudfactory.com

Installation

Claude Desktop

{
  "mcpServers": {
    "ead-factory": {
      "args": [
        "-y",
        "@g-digital/mcp-ead-factory"
      ],
      "command": "npx",
      "env": {
        "MCP_SVC_CLIENT_ID": "",
        "MCP_SVC_CLIENT_SECRET": "<PASTE_MCP_SVC_CLIENT_SECRET_HERE>",
        "MCP_SVC_TOKEN_URL": ""
      }
    }
  }
}

Need credentials? See: https://digitaltrust.gcloudfactory.com

Claude Code (CLI)

{
  "mcpServers": {
    "ead-factory": {
      "args": [
        "-y",
        "@g-digital/mcp-ead-factory"
      ],
      "command": "npx",
      "env": {
        "MCP_SVC_CLIENT_ID": "",
        "MCP_SVC_CLIENT_SECRET": "<PASTE_MCP_SVC_CLIENT_SECRET_HERE>",
        "MCP_SVC_TOKEN_URL": ""
      }
    }
  }
}

Need credentials? See: https://digitaltrust.gcloudfactory.com

Cursor

{
  "mcpServers": {
    "ead-factory": {
      "args": [
        "-y",
        "@g-digital/mcp-ead-factory"
      ],
      "command": "npx",
      "env": {
        "MCP_SVC_CLIENT_ID": "",
        "MCP_SVC_CLIENT_SECRET": "<PASTE_MCP_SVC_CLIENT_SECRET_HERE>",
        "MCP_SVC_TOKEN_URL": ""
      }
    }
  }
}

Need credentials? See: https://digitaltrust.gcloudfactory.com

Windsurf

{
  "mcpServers": {
    "ead-factory": {
      "args": [
        "-y",
        "@g-digital/mcp-ead-factory"
      ],
      "command": "npx",
      "env": {
        "MCP_SVC_CLIENT_ID": "",
        "MCP_SVC_CLIENT_SECRET": "<PASTE_MCP_SVC_CLIENT_SECRET_HERE>",
        "MCP_SVC_TOKEN_URL": ""
      }
    }
  }
}

Need credentials? See: https://digitaltrust.gcloudfactory.com

Cline

{
  "mcpServers": {
    "ead-factory": {
      "args": [
        "-y",
        "@g-digital/mcp-ead-factory"
      ],
      "command": "npx",
      "env": {
        "MCP_SVC_CLIENT_ID": "",
        "MCP_SVC_CLIENT_SECRET": "<PASTE_MCP_SVC_CLIENT_SECRET_HERE>",
        "MCP_SVC_TOKEN_URL": ""
      }
    }
  }
}

Need credentials? See: https://digitaltrust.gcloudfactory.com

VS Code

{
  "servers": {
    "ead-factory": {
      "args": [
        "-y",
        "@g-digital/mcp-ead-factory"
      ],
      "command": "npx",
      "env": {
        "MCP_SVC_CLIENT_ID": "",
        "MCP_SVC_CLIENT_SECRET": "<PASTE_MCP_SVC_CLIENT_SECRET_HERE>",
        "MCP_SVC_TOKEN_URL": ""
      }
    }
  }
}

Need credentials? See: https://digitaltrust.gcloudfactory.com

JetBrains

{
  "mcpServers": {
    "ead-factory": {
      "args": [
        "-y",
        "@g-digital/mcp-ead-factory"
      ],
      "command": "npx",
      "env": {
        "MCP_SVC_CLIENT_ID": "",
        "MCP_SVC_CLIENT_SECRET": "<PASTE_MCP_SVC_CLIENT_SECRET_HERE>",
        "MCP_SVC_TOKEN_URL": ""
      }
    }
  }
}

Need credentials? See: https://digitaltrust.gcloudfactory.com

Zed

{
  "mcpServers": {
    "ead-factory": {
      "args": [
        "-y",
        "@g-digital/mcp-ead-factory"
      ],
      "command": "npx",
      "env": {
        "MCP_SVC_CLIENT_ID": "",
        "MCP_SVC_CLIENT_SECRET": "<PASTE_MCP_SVC_CLIENT_SECRET_HERE>",
        "MCP_SVC_TOKEN_URL": ""
      }
    }
  }
}

Need credentials? See: https://digitaltrust.gcloudfactory.com

Claude Desktop / Claude Code

Add to your ~/.claude.json or claude_desktop_config.json:

{
  "mcpServers": {
    "ead-factory": {
      "command": "npx",
      "args": ["-y", "@g-digital/mcp-ead-factory"],
      "env": {
        "MCP_SVC_TOKEN_URL": "https://your-idp.example.com/oauth2/v1/token",
        "MCP_SVC_CLIENT_ID": "your-client-id",
        "MCP_SVC_CLIENT_SECRET": "your-client-secret"
      }
    }
  }
}

Docker

docker run --rm -i \
  -e MCP_SVC_TOKEN_URL=https://your-idp.example.com/oauth2/v1/token \
  -e MCP_SVC_CLIENT_ID=your-client-id \
  -e MCP_SVC_CLIENT_SECRET=your-client-secret \
  gdigital/ead-factory:latest

Environment Variables

Name	Required	Secret	Description
MCP_ALLOW_INSECURE_FILE_URL	No	No	Set to "true" to allow plain http:// fileUrl downloads in evidence_create (default https-only). Private/internal addresses are rejected regardless (resolution-time check; see the documented DNS TOCTOU limitation in the hosted-deployment runbook).
MCP_ALLOW_UNVERIFIED_BEARER	No	No	Escape hatch for MCP_HTTP_PUBLIC=true WITHOUT inbound-token introspection: set to "true" ONLY when an upstream gateway already verifies Bearer tokens. The server logs a prominent warning and forwards tokens upstream unverified.
MCP_ALLOWED_HOSTS	No	No	Comma-separated allowed Host headers. Empty = Host validation disabled (default). When set, requests with a Host outside the list are rejected.
MCP_ALLOWED_ORIGINS	No	No	Comma-separated allowed browser Origins (DNS-rebinding defense). Empty = reject any request carrying an Origin header; non-browser clients (CLI/SDK) send no Origin and are always allowed. Use '*' to allow all.
MCP_API_BASE_URL	No	No	Gateway ROOT URL (e.g. https://api.int.gcloudfactory.com) — each manager's path prefix is appended automatically; do NOT include a manager path here
MCP_API_BASE_URL_CHAT	No	No	Full base URL override for the chat manager, used as-is (default: https://api.gcloudfactory.com/chat-bot-manager)
MCP_API_BASE_URL_EVIDENCE	No	No	Full base URL override for the evidence manager, used as-is (default: https://api.gcloudfactory.com/digital-trust)
MCP_API_BASE_URL_NOTIFICATION	No	No	Full base URL override for the notification manager, used as-is (default: https://api.gcloudfactory.com/notifications)
MCP_API_BASE_URL_SIGNATURE	No	No	Full base URL override for the signature manager, used as-is (default: https://api.gcloudfactory.com/signature-manager)
MCP_HTTP_HOST	No	No	Interface the HTTP transport binds to. Default 127.0.0.1 (localhost only). Set 0.0.0.0 to expose on all interfaces (containers do this automatically).
MCP_HTTP_MAX_BODY_BYTES	No	No	Maximum accepted POST /mcp request-body size in bytes (default 16777216 = 16 MiB — sized so base64 file uploads within the documented tool limits fit). Oversized requests get a 413 JSON-RPC error before/while reading — closes a memory-exhaustion DoS vector in public deployments. Note: base64 file sources are capped by this limit BEFORE MCP_FILE_MAX_BYTES applies.
MCP_HTTP_PUBLIC	No	No	Set to "true" for public/multi-tenant deployments. Activates Host validation and refuses to start unless (1) MCP_ALLOWED_ORIGINS or MCP_ALLOWED_HOSTS is set AND (2) inbound Bearer introspection is configured (MCP_SVC_INTROSPECT_URL + MCP_SVC_CLIENT_ID/SECRET) or MCP_ALLOW_UNVERIFIED_BEARER=true is set explicitly (fail-closed).
MCP_SVC_CLIENT_ID	Yes	No	OAuth2 client_credentials client ID.
MCP_SVC_CLIENT_SECRET	Yes	Yes	OAuth2 client_credentials client secret. (See https://digitaltrust.gcloudfactory.com for credential acquisition.)
MCP_SVC_INTROSPECT_URL	No	No	RFC 7662 token introspection URL for inbound Bearer verification in HTTP mode. Opt-in, and required when MCP_HTTP_PUBLIC=true. Leave empty for stdio (local) use.
MCP_SVC_SCOPE	No	No	Optional OAuth2 scope for the service-account token request.
MCP_SVC_TOKEN_URL	Yes	No	Token endpoint URL for the OAuth2 client_credentials flow.
PORT	No	No	HTTP port when running in hosted (HTTP) mode; ignored in stdio mode

Variable	Required	Description
MCP_SVC_TOKEN_URL	Required	OAuth2 client_credentials token endpoint URL
MCP_SVC_CLIENT_ID	Required	OAuth2 client ID
MCP_SVC_CLIENT_SECRET	Required	OAuth2 client secret
MCP_SVC_SCOPE	Optional	OAuth2 scope for the token request
MCP_AUTH_JWT	Optional	Pre-seeded JWT (skips interactive login)
MCP_OTEL_ENABLED	Optional	Set to true to enable OpenTelemetry tracing
MCP_API_BASE_URL	Optional	Gateway root URL — each manager's path prefix is appended automatically; use MCP_API_BASE_URL_<MANAGER> for a full per-manager override (used as-is)

Bundled Skills

This package ships Claude Code slash-commands under .claude/commands/. After install, invoke them from Claude Code:

• /create-internal-evidence — step-by-step workflow guide
• /create-signature-request — step-by-step workflow guide
• /create-notification-request — step-by-step workflow guide

See docs/agent-prompts.md for end-to-end prompt examples and the tool sequences they trigger.

Prefer to code against the REST API directly?

You don't have to go through this MCP server. This repo also ships a Claude Code skill — a standalone integration guide (authentication, call ordering, options, enums, gotchas) for programming directly against the REST API:

• Skill: .claude/skills/ead-factory-api/SKILL.md — open this repo in Claude Code and it is available directly.
• As a reference: docs/api-integration-skill.md.

It is independent of the MCP tools and the n8n node — pick whichever entry point fits your integration.

Available Tools

This server exposes 64 tools:

Tool	Description
evidence_case_file_delete_bulk	Permanently deletes SEVERAL case files in one call. Destructive and not reversible — confirm the caseFileIds first (evidence_case_file_search) and prefer closing a case file via evidence_case_file_status_update when you only need to end work on it.
evidence_case_file_search	Searches case files with optional filters (name, status, dates, pagination). Use to find a caseFileId when you don't have it, or to list what exists before creating a new case file with evidence_case_file_create.
evidence_case_file_create	Creates a new case file — the top-level container for evidence groups, evidence, and reports. Use this first, before any other Evidence-manager tool. No prerequisites. Unlike most create tools, id is CALLER-SUPPLIED (generate a fresh UUID v4 yourself, this API does not assign one) — there is no name field, use title instead. Example: evidence_case_file_create({ id: '<generate a UUID v4>', title: 'Q1 2026 audit' }) — use that same id as caseFileId in evidence_group_create.
evidence_case_file_update_bulk	Updates the editable metadata of SEVERAL case files in one call (each entry carries its own caseFileId and changes). Prefer evidence_case_file_update for a single case file. Requires: the caseFileIds of every case file to update (evidence_case_file_search to find them).
evidence_case_file_get	Retrieves a case file's details by id. Use to confirm a case file exists, or to check its status before creating evidence groups or reports under it. Requires: evidence_case_file_create → caseFileId.
evidence_case_file_update	Updates a single case file's editable metadata (e.g. name, description). Requires: evidence_case_file_create or evidence_case_file_search → caseFileId. To change its OPEN/CLOSE lifecycle state use evidence_case_file_status_update instead.
evidence_case_file_relationship_assign	Assigns a relationship on a case file, linking it to an external/related entity reference for traceability. Use after creating the case file when your workflow tracks which matter, client, or system record it belongs to. Requires: evidence_case_file_create → caseFileId.
evidence_case_file_report_preview	Generates an UNSIGNED preview of a case file's evidentiary report so you can check its content before producing the signed version with evidence_case_file_report_generate. Requires: evidence_case_file_create → caseFileId, with sealed evidence groups inside.
evidence_case_file_report_generate	Generates the SIGNED evidentiary report of a case file (the certified document covering its sealed evidence). Returns a reportId — then call evidence_case_file_report_pdf_url_get (PDF) or evidence_case_file_report_zip_url_get (full package) to download. Requires: evidence_case_file_create → caseFileId; preview first with evidence_case_file_report_preview.
evidence_case_file_report_update	Updates a previously generated case-file report's metadata. Requires: evidence_case_file_report_generate → reportId (and the caseFileId it belongs to). Does not re-generate the report content — generate a new report for updated evidence.
evidence_case_file_report_pdf_url_get	Retrieves a download URL for the signed PDF document of a case-file report. Requires: evidence_case_file_report_generate → reportId (plus caseFileId). For the full evidence package (PDF + files + verification data) use evidence_case_file_report_zip_url_get.
evidence_case_file_report_zip_url_get	Retrieves a download URL for the complete ZIP package of a case-file report (signed PDF plus evidence files and verification material). Requires: evidence_case_file_report_generate → reportId (plus caseFileId).
evidence_case_file_status_update	Changes a case file's lifecycle status (OPEN or CLOSE). Use CLOSE when work on the case file is finished and it should no longer accept new evidence groups; reopen with OPEN. Requires: evidence_case_file_create or evidence_case_file_search → caseFileId.
evidence_report_delete	Permanently deletes a generated report by its reportId. Destructive — the signed document and its package stop being downloadable. Requires: evidence_case_file_report_generate → reportId. The underlying case file and its evidence are NOT touched.
evidence_group_evidence_register	Register a new evidence in a group — the first step of a 2-step upload (register, then PUT the file bytes to the returned upload URL). Use for each file you want to add to an OPEN evidence group. Requires: evidence_group_create → evidenceGroupId, evidence_case_file_create → caseFileId, and the file's SHA-256 hash (compute it before calling, or use the evidence_create_sealed composite tool which does this for you). Do not call evidence_group_close until every registered evidence's bytes have been uploaded. Custody type: INTERNAL = EAD Factory stores and custodies the file itself.
evidence_group_evidence_get	Gets one evidence's details (metadata, hash, custody, upload state) inside a specific evidence group. Requires: evidence_case_file_create → caseFileId, evidence_group_create → evidenceGroupId, evidence_group_evidence_register → evidenceId.
evidence_group_evidence_delete	Permanently deletes one evidence from an evidence group. Only makes sense while the group is still OPEN (sealed groups are immutable). Requires: caseFileId + evidenceGroupId + evidence_group_evidence_register → evidenceId.
evidence_group_evidence_download_url_create	Creates a temporary download URL for an evidence file stored in a group (INTERNAL custody). Use to retrieve the original bytes after upload. Requires: caseFileId + evidenceGroupId + evidence_group_evidence_register → evidenceId.
evidence_group_evidence_upload_url_create	Creates a fresh presigned upload URL for an ALREADY-REGISTERED evidence in a group — step 2 of the register-then-upload flow (PUT the file bytes to the returned URL). Use when the URL from evidence_group_evidence_register expired or was lost. Requires: caseFileId + evidenceGroupId + evidenceId. evidence_create_sealed does all of this automatically.
evidence_thumbnail_url_get	Retrieves a temporary URL for an image evidence's thumbnail at the requested size — use for quick visual verification without downloading the original file. Requires: caseFileId + evidenceGroupId + evidenceId (image-type evidence only) and a thumbnailSize path value.
evidence_search	Searches evidences across the tenant with optional filters (state, dates, pagination) — not scoped to one group. Use to find an evidenceId or audit what exists; for one group's content use evidence_group_get instead.
generate_evidence	Register a new evidence (legacy top-level name; not scoped to a specific evidence group at creation time, unlike evidence_group_evidence_register). Use for standalone evidence outside the group-based flow, or for continuity with the legacy EAD-Factory-MCP integration. Requires the file's SHA-256 hash. Prefer evidence_create_sealed for a new, complete evidence group + evidence + seal flow. Custody type: INTERNAL = EAD Factory stores and custodies the file itself.
evidence_update_bulk	Updates the editable metadata of SEVERAL evidences in one call (each entry carries its own evidenceId and changes). Prefer evidence_update for a single evidence. Requires: the evidenceIds to update (evidence_search to find them).
get_evidence	Retrieves an evidence record's details and status by id (legacy top-level name). Use to check an evidence's timestamping status (IN_PROCESS / COMPLETED / ERROR) after registration. Requires: generate_evidence or evidence_group_evidence_register → evidenceId.
evidence_delete	Permanently deletes one evidence by evidenceId (no group path needed). Destructive; sealed content should not be deleted — use only for drafts or mis-registered evidence. Requires: evidence_group_evidence_register or generate_evidence → evidenceId.
evidence_update	Updates a single evidence's editable metadata (e.g. name, description) by evidenceId. The file content and hash are immutable — re-register a new evidence for changed files. Requires: evidence_group_evidence_register or generate_evidence → evidenceId.
evidence_download_url_create	Creates a temporary download URL for an evidence's stored file when you only hold the evidenceId (no group path needed — same bytes as evidence_group_evidence_download_url_create). Requires: generate_evidence or evidence_group_evidence_register → evidenceId.
evidence_upload_url_create	Creates a fresh presigned upload URL for an already-registered evidence by evidenceId — step 2 of the register-then-upload flow (PUT the file bytes to the returned URL). Pairs with generate_evidence the way evidence_group_evidence_upload_url_create pairs with evidence_group_evidence_register. Requires: generate_evidence → evidenceId.
evidence_temp_file_upload_url_create	Creates a presigned upload URL for a TEMPORARY file not yet registered as evidence. Use for staging content that another operation will reference; for real evidence prefer the register-then-upload flow (evidence_group_evidence_register) or the evidence_create_sealed composite, which handle registration and upload together.
evidence_multipart_upload_start	Starts a MULTIPART upload session for a large evidence file (returns the upload coordinates for uploading the file in parts). Use when a single presigned PUT is not enough for the file size. Requires: an already-registered evidence (evidence_group_evidence_register or generate_evidence → evidenceId) and the file name.
evidence_group_delete_bulk	Permanently deletes SEVERAL evidence groups of one case file in one call. Destructive — prefer evidence_group_discard for a single group, and never delete sealed groups that back issued reports. Requires: caseFileId + the evidenceGroupIds to delete (evidence_group_search to find them).
evidence_group_create	Creates an evidence group inside a case file — evidence records are always registered inside a group, never standalone (except via the legacy generate_evidence tool). Use when starting a new batch of related evidence (e.g. all files for one incident). Like evidence_case_file_create, id is CALLER-SUPPLIED (generate a fresh UUID v4 yourself). Requires: evidence_case_file_create → caseFileId. Example: evidence_group_create({ id: '<generate a UUID v4>', caseFileId, type: 'FILE' }) returns { status: 'OPEN', ... } — use that same id as evidenceGroupId in evidence_group_evidence_register, then evidence_group_close once every evidence in the group has been uploaded.
evidence_group_get	Gets one evidence group's details — status (OPEN/CLOSED), seal information, and its evidences with their upload states. Use to verify every file was uploaded before evidence_group_close, or to check the seal after closing. Requires: caseFileId + evidence_group_create → evidenceGroupId.
evidence_group_discard	Discards (deletes) an OPEN evidence group and its registered evidences — use to abandon a group you no longer intend to seal. Destructive; sealed (CLOSED) groups are immutable evidence and should not be discarded. Requires: caseFileId + evidence_group_create → evidenceGroupId.
evidence_group_update	Updates an evidence group's editable metadata (e.g. name, description). Only meaningful while the group is OPEN — sealed groups are immutable. Requires: caseFileId + evidence_group_create → evidenceGroupId.
evidence_group_close	Seals (closes) an evidence group, triggering qualified timestamping — after this, no more evidence can be added. Use only after every evidence registered in the group has had its file bytes uploaded to the presigned URL from evidence_group_evidence_register. Requires: evidence_group_create → evidenceGroupId, evidence_case_file_create → caseFileId, and the current evidencesCount. ASYNC: the group transitions OPEN → CLOSING → CLOSED; poll evidence_group_get until status is CLOSED before generating a report. Prefer the evidence_create_sealed composite tool for a new group — it registers, uploads, closes, and waits for CLOSED in one call.
evidence_group_search	Searches evidence groups across case files with optional filters (status, dates, pagination). Use to find an evidenceGroupId or list groups pending sealing; for one group's full detail use evidence_group_get.
evidence_group_update_bulk	Updates the editable metadata of SEVERAL evidence groups in one call (each entry carries its own evidenceGroupId and changes). Prefer evidence_group_update for a single group. Requires: the evidenceGroupIds to update (evidence_group_search to find them).
evidence_report_pdf_url_get	Retrieves a download URL for a report's signed PDF when you only hold the reportId (no caseFileId path needed — same document evidence_case_file_report_pdf_url_get returns). Requires: evidence_case_file_report_generate → reportId.
evidence_report_zip_url_get	Retrieves a download URL for a report's complete ZIP package when you only hold the reportId (same package evidence_case_file_report_zip_url_get returns). Requires: evidence_case_file_report_generate → reportId.
create_signature_request	Creates a new signature request — the top-level container for documents and signatories in a signing flow. Use this first, before add_document_to_signature_request. No prerequisites. Example: create_signature_request({ name: 'NDA — Acme Corp', createdBy: 'jane@company.com' }) returns { id, ... } — use the returned id as signatureRequestId in subsequent calls.
add_document_to_signature_request	Adds a document to a signature request — the first step of a 2-step upload (add, then PUT the file bytes to the returned upload URL). Use once per document that needs signing. Requires: create_signature_request → signatureRequestId, and the document's SHA-256 hash (compute it before calling, or use the signature_request_full composite tool). Add signatories with add_signatory_to_document before activating. Signature type: INTERPOSITION = EAD Factory mediates the signing act on the signatory's behalf (e.g. an OTP sent via WhatsApp/SMS) — the signatory needs no software or certificate.
add_signatory_to_document	Adds a signatory (by name + email) to a document within a signature request. Use once per person who needs to sign that specific document. Requires: create_signature_request → signatureRequestId, add_document_to_signature_request → documentId. Add all signatories before calling activate_signature_request — signatories cannot be added after activation.
add_validator_to_signatory	Adds a validator to a document's signatory: a person who must approve (validate) that signatory's identity or the document before the signature can proceed. Add validators BEFORE activate_signature_request. Requires: add_document_to_signature_request → documentId, add_signatory_to_document → signatoryId. The signature_request_full composite accepts validators inline.
add_observer_to_document	Adds an observer to a signature request's document: a person who receives read-only visibility of the process and the signed result without signing. Add observers BEFORE activate_signature_request. Requires: create_signature_request → signatureRequestId, add_document_to_signature_request → documentId.
activate_signature_request	Activates a signature request, sending signing notifications to every added signatory — after this, no more documents or signatories can be added (use add_validator_to_signatory / add_observer_to_document before activating if you need those roles). Requires: create_signature_request → signatureRequestId, at least one document (add_document_to_signature_request) with its bytes uploaded, and at least one signatory per document (add_signatory_to_document). Prefer signature_request_full for a new request — it creates, adds documents+signatories, and activates in one call.
get_signature_request	Retrieves a signature request's details and status by id (legacy top-level name). Use to check signing progress after activation. Requires: create_signature_request → signatureRequestId.
signature_request_list	List signature requests, optionally filtered by close condition among other criteria. Use to find a request when you don't have its id (e.g. by name or status). Signature-request close condition: ALL_REQUIRED = The request only closes once EVERY signatory has signed.
signature_request_cancel	Cancels an active signature request — no further signing can occur. Use when a request was activated in error or is no longer needed. Requires: create_signature_request → signatureRequestId. Cannot be undone.
signature_certificate_generate	Generates the well-signed appearance certificate document for a signature request. Use once every signatory has signed (check with get_signature_request). Requires: create_signature_request → signatureRequestId. Unlike EAD Enterprise's equivalent tool (a GET that polls an already-generated URL), this triggers generation and returns the result in the same call.
signature_coordinate_set	Sets the on-page (x, y, page) coordinates where a signatory's signature appears on a document. Use after add_signatory_to_document if the signature placement needs to be explicit rather than auto-positioned. Requires: create_signature_request → signatureRequestId, add_document_to_signature_request → documentId, add_signatory_to_document → signatoryId.
signature_document_download_url_get	Retrieves the download link for the FINAL, well-signed version of a document — the actual signed file, once every signatory has signed. Requires: create_signature_request → signatureRequestId, add_document_to_signature_request → documentId. Use get_signature_request first to confirm signing is complete before calling this. Not to be confused with signature_certificate_generate, which produces a separate appearance certificate document, not the signed document itself.
notification_request_create	Creates a new notification request (draft) — the top-level container for receivers and attachments in a certified-notification flow. Use this first, before any other Notification-manager tool. No prerequisites. Set autosend: true to also send immediately on activation of the first receiver batch, or leave it false to add receivers/documents over several calls before calling notification_request_send yourself. Returns requestId and one notificationId per receiver already on the request (if any were included inline).
notification_request_send	Activates a notification request, triggering delivery to every added receiver across their configured channels. Runs as an MCP Task (bounded-polling until every receiver's notification leaves its in-flight state — EAD Factory's upstream emits no events for this transition). Requires: notification_request_create → requestId, at least one receiver (notification_receiver_add). Use notification_request_status to check progress without waiting for the Task to complete.
notification_request_status	Searches notifications, optionally filtered by requestId and/or delivery state — the closest equivalent to "checking a request's status": a request fans out into one notification per receiver, each with its own state history, so this returns every notification matching the filter with its current state. Omit requestId to search across every request. Requires: notification_request_create → requestId (if filtering to one request).
notification_receiver_add	Adds one or more receivers to a draft notification request — each receiver becomes its own per-channel notification under the shared requestId (there is no separate receiverId; track receivers by the notificationIds this call returns). Use once per batch of receivers before calling notification_request_send. Requires: notification_request_create → requestId. Each receiver's shape depends on its provider (SMTP/SMS/NOTICEMAN/WFB) — see the schema for the fields each variant needs. Notification delivery provider: SMTP = Email delivery.
notification_document_add	Registers one or more document attachments (metadata only — fileName + SHA-256 hash) on a draft notification request. Use once per batch of documents before activating. Requires: notification_request_create → requestId. This only registers metadata; call notification_document_upload_url_create for each registered attachment to get a URL to PUT the actual file bytes to.
notification_document_upload_url_create	Creates a presigned upload URL for a document attachment already registered via notification_document_add — PUT the file bytes to the returned URL. Requires: notification_request_create → requestId, notification_document_add → attachmentId.
notification_document_download_url_create	Creates a presigned download URL for a document attachment on a notification request. Requires: notification_request_create → requestId, notification_document_add → attachmentId.
notification_certificate_generate	Generates a delivery-certificate report for one or more notifications on a request. Use once notification_request_status shows the relevant notifications have left their in-flight state. Requires: notification_request_create → requestId, notification_request_status → notificationIds. Returns a reportId — call notification_certificate_pdf_url_get with it to get the actual download URL (2-step, same shape as Evidence's report tools).
notification_certificate_pdf_url_get	Retrieves the download URL for a previously generated notification delivery certificate. Requires: notification_certificate_generate → reportId.
evidence_create_sealed	Creates (or reuses) an evidence group, registers and uploads one or more files as evidence, then seals (closes) the group — the full flagship evidence flow in one call instead of evidence_group_create + N x (evidence_group_evidence_register + upload) + evidence_group_close. Bounded-polls until the group reaches CLOSED (EAD Factory's upstream emits no events for this transition): task-aware MCP clients get an asynchronous Task; any other client simply receives the final result synchronously. Requires: case_file_create -> caseFileId. Provide evidenceGroupId to add to an existing OPEN group instead of creating one. Each evidence's file uses the shared FileInput contract (local path, base64, https URL, or n8n binary item) — never a bespoke file field. On success returns the group id, its final CLOSED status, and each evidence's id + sha256. Use the atomic evidence_group_create / evidence_group_evidence_register / evidence_group_close tools instead when you need to inspect or react to each intermediate step.
signature_request_full	Creates a signature request, adds one or more documents (each with its own signatories), and activates it — the full flagship signature flow in one call instead of create_signature_request + N x (add_document_to_signature_request + add_signatory_to_document) + activate_signature_request. Each document's file uses the shared FileInput contract (local path, base64, https URL, or n8n binary item) — never a bespoke file field. Set activate: false to leave the request in draft so you can add validators/observers (add_validator_to_signatory, add_observer_to_document) before activating it yourself. On success returns the request id, each document's id, and each signatory's id. Use the atomic tools instead when you need to inspect or react to each intermediate step, or need validators/observers before activation.
ead_factory_help	Returns an overview of EAD Factory's managers (Evidence, Signature, Notification, Chat), their key tools and starting points, and guidance on when to use a composite workflow tool (e.g. evidence_create_sealed) versus the atomic tools. Call this first if you're unsure which tool to use — no credentials needed.

MCP Tasks support

notification_request_send runs as an MCP Task and requires a client with MCP Tasks support — from other clients the call fails; instead, create the request with autosend: true (notification_request_create) and poll notification_request_status yourself.

evidence_create_sealed works from any client: task-aware clients get an asynchronous Task; all others simply receive the final result synchronously.

Coexistence

This MCP server is the current, actively maintained interface for the EAD Factory (Digital Trust) platform.

It coexists safely with any other MCP servers in your setup — it exposes only EAD-Factory-namespaced tools and shares no local state with other servers.

License

MIT — see LICENSE.