Ead Factory Mcp

Name: Ead Factory Mcp
Author: g-digital-by-Garrigues

@g-digital-by-Garrigues

关于 Ead Factory Mcp

EAD Factory MCP — Digital Trust services APIs for your agents. Bridges any MCP-compatible client (Claude Code, Claude Desktop, Cursor, etc.) to Garrigues' Evidence Manager + Signature Manager APIs.

#digital-trust #e-signature #evidence #eidas #legal-tech #qualified-trust #garrigues #ead-trust #okta-auth #signature-workflow

基本信息

分类

开发工具

许可证

MIT

运行时

node

传输方式

stdio

发布者

g-digital-by-Garrigues

提交者

Hugo Alonso

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "ead-factory": {
      "command": "npx",
      "args": [
        "-y",
        "@g-digital/mcp-ead-factory"
      ],
      "env": {
        "OKTA_TOKEN_URL": "https://<your-okta-tenant>.okta.com/oauth2/<authServer>/v1/token",
        "OKTA_CLIENT_ID": "<your-okta-client-id>",
        "OKTA_CLIENT_SECRET": "<your-okta-client-secret>",
        "OKTA_SCOPE": "token",
        "API_BASE_URL": "https://api.int.gcloudfactory.com/digital-trust",
        "SIGNATURE_API_BASE_URL": "https://api.int.gcloudfactory.com/signature-manager"
      }
    }
  }
}

工具

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.

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.

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.

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).

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.

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.

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.

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.

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.

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.

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.

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).

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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).

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.

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.

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.

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.

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.

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.

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.

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).

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.

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.

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.

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.

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.

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.

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).

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.

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.

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: '[email protected]' }) returns { id, ... } — use the returned id as signatureRequestId in subsequent calls.

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.

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.

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.

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.

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.

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.

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.

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.

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.

概览

What is EAD Factory MCP Server?

EAD Factory MCP Server is a Model Context Protocol server that provides qualified evidence, signature, notifications, and chat capabilities to AI agents. It integrates with the EAD Factory digital trust platform.

How to use EAD Factory MCP Server?

Install via npx -y @g-digital/mcp-ead-factory or Docker, then configure the required environment variables (MCP_SVC_TOKEN_URL, MCP_SVC_CLIENT_ID, MCP_SVC_CLIENT_SECRET). Credentials are obtained from https://digitaltrust.gcloudfactory.com. Preconfigured examples are provided for Claude Desktop, Claude Code, Cursor, Windsurf, Cline, VS Code, JetBrains, and Zed. Bundled Claude Code slash-commands (/create-internal-evidence, etc.) offer workflow guides.

Key features of EAD Factory MCP Server

64 tools for evidence, signature, notifications, and chat
Available via npm, Docker, Smithery, and n8n
Bundled Claude Code slash-command workflows
OAuth2 client_credentials authentication
Configurable HTTP transport with security features
Docker image for containerized deployment

Use cases of EAD Factory MCP Server

Automate evidence creation and case file management
Send qualified signature requests via AI agents
Trigger notification requests through AI chat workflows
Search and manage evidence records programmatically
Manage digital trust processes without manual intervention

FAQ from EAD Factory MCP Server

What credentials are required?

You need OAuth2 client credentials (client ID, client secret, and token URL). Obtain them from https://digitaltrust.gcloudfactory.com.

Which environment variables are mandatory?

MCP_SVC_TOKEN_URL, MCP_SVC_CLIENT_ID, and MCP_SVC_CLIENT_SECRET are required. Optional variables allow customizing API base URLs, HTTP settings, and