mcpbeat Sign in

Emboss MCP Server

by edwinorange Your server? Claim it
answering

Emboss is answering right now. Last checked 3 min ago. It exposes 28 tools. Last commit 19 Oct 2026.

Make any PDF form fillable, fill it from data or documents, read it back, and fax the result.

Uptime history 58 days of history · worst day 95%
58 days agonow
100.0%
Uptime 24h
91 of 91 checks
28
Tools
read from the server
339 ms
Response time
average over 24h
0
Stars
last commit 19 Oct 2026

What changed 16

Every tool that appeared, vanished or quietly changed what it asks for. Recorded since 12 October 2026. No other catalogue keeps this.

19 Oct 8 tool descriptions were rewritten compose_pdf, delete_pages, extract_pages and 5 more
19 Oct a tool changed version2 times that day
19 Oct a tool appeared flatten_pdf
17 Oct 3 tool descriptions were rewritten get_fax, get_usage, send_fax
17 Oct a tool changed version
12 Oct a tool changed version
and 1 more, back to 12 October 2026

Nothing serious here today

Today is the operative word: we check Emboss every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.

Three servers free · no card

Connect this server

Endpoint below is the one we actually reach during checks — not the one copied from a README. Last verified 3 min ago.

run in your terminal
claude mcp add emboss --transport http https://api.getemboss.ai/mcp
Settings → Connectors → Add custom connector, then paste this address
https://api.getemboss.ai/mcp
~/.codex/config.toml
[mcp_servers.emboss]
url = "https://api.getemboss.ai/mcp"
.cursor/mcp.json
{
  "mcpServers": {
    "emboss": {
      "url": "https://api.getemboss.ai/mcp"
    }
  }
}
.vscode/mcp.json
{
  "servers": {
    "emboss": {
      "type": "http",
      "url": "https://api.getemboss.ai/mcp"
    }
  }
}

Available tools 28

Read directly from the server with tools/list, grouped by what they act on. If a tool disappears, we record the date.

form
create_form
Turn a flat or scanned PDF form into a fillable PDF with detected fields. Upload a PDF (pdf_url: public https link, or pdf_base64) and detect its fields. Returns when detection finishes (usually under two minutes). Give library=<slug> from find_form to start from a form Emboss already keeps. Billed as one form creation; the first 5 each month are free. Optional retention: ephemeral or account_default (default: the account's own setting). Ephemeral processing: source and output documents are deleted 60 to 70 minutes after the last activity on a form.
delete_form
Permanently removes a form and its PDFs from the user's library.
find_form
Find a blank government form in the Emboss library by name or number, ready to fill. Search the library of blank US federal forms by name, number or alias (w9, W-9, form w 9 and taxpayer identification all find the same form). Returns each match's slug, title, agency, revision and page count. Send the whole query once, and do not search as the user types: each search spends one of the account's rate-limit slots. Pass a slug as create_form's library argument to start from that form. Free.
get_form
Status of one form. When ready, returns every field (id, label, kind, options, required) and a download link for the fillable PDF.
read_form
Read the values out of a filled PDF form, labelled by field, with a completeness verdict when the form is known. Give the filled PDF (pdf_url: public https link, or pdf_base64) and, when you know it, form_id. Returns every form box in the file with its value and whether it is filled; when the form is recognised, each value also carries its label and whether it was required, plus a completeness verdict. Nothing is stored: read-back never keeps your file. Billed as one read; the first 5 each month are free.
fill
fill_batch
Fill one copy of a PDF form for every row of a spreadsheet. Fill a form for every row of a CSV (csv_text or csv_url, up to 1000 rows). mapping is {column: field_id}; omit it to use the suggested mapping. If mapping is omitted, suggest_mapping runs first and is billed as one context fill. Batches up to 50 rows finish before this returns; larger ones keep running, poll get_batch. Each row is billed as one fill. retention (ephemeral or account_default) is accepted but changes nothing: the form keeps the setting it was created with.
fill_form
Fill a PDF form from structured data such as JSON or key-value answers. Fill a ready form with values keyed by field id or field label. Checkboxes take yes or no; choice fields need one of their options. To lock the filled PDF, call flatten_pdf with its artifact_id; fill_form itself never flattens. Returns a download link plus any labels that did not match (ask the user about those). Billed as one fill; the first 5 each month are free. retention (ephemeral or account_default) is accepted but changes nothing: the form keeps the setting it was created with.
fill_form_from_context
Fill a PDF form using facts found in supporting documents. Emboss reads the answers out of context_text and/or context_urls (public https links to PDFs, Word, spreadsheets, images, or text) and fills the form. Give form_id for a form already in the library, or pdf_url/pdf_base64 for a new one. Give library=<slug> from find_form to start from a form Emboss already keeps. policy is safe (write high and medium confidence) or strict (high only). Returns a job_id; poll get_job about every 20 seconds. Billed as one context fill; the first 5 each month are free. Optional retention: ephemeral or account_default (default: the account's own setting); it applies only when this call creates the form (pdf_url, pdf_base64 or library), since an existing form keeps the setting it was created with. Ephemeral processing: source and output documents are deleted 60 to 70 minutes after the last activity on a form.
attachment
add_attachment
Prepare a form from documents: a proposal of every answer with its source, what is missing, and where documents disagree, before writing, plus attachments for a package. Attach one document (file_url: public https link, or file_base64) to a proposal from prepare_form. PDF, PNG and JPEG only, at most 20 per proposal. requirement: the name of an entry in the form's attachments_required list to satisfy it; uploading a second document for the same requirement replaces the first. Free: attachments are part of the already-billed prepare.
batch
get_batch
Progress of a fill_batch run with a download link per finished row (first 100) and a zip link when done.
commit
commit_proposal
Write or confirm the answers, render the form with text that fits, check it, and return the PDF with a report and a receipt, or one submission package. Give the proposal_id from prepare_form; confirm is a list of field ids whose proposed value is accepted as it is (the ids get_proposal lists under confirmations); plus any values the agent decided for fields still open (field_id, value, and an optional note), and a policy (safe or strict) to override the proposal's own. Set package to true to also receive one submission package: the filled form, the attachments added to this proposal in the order the form asks for them, and the receipt embedded in the file; that is billed as one package. Returns a job_id; poll get_job about every 20 seconds for the download link, report, receipt and, when asked for, package_url. Free within the proposal's already-billed prepare, up to 10 commits per proposal. Checkboxes take yes or no; yes ticks that box. In a Yes/No pair, say yes to the box you want; two yeses in one pair are refused.
compose
compose_pdf
Free. Build one PDF from pages of several artifacts, in the order and rotation you name. Build one PDF from several artifacts: sources is a list of {artifact_id, pages, rotate}; pages is a selection like 1-3,7 (omit for every page), rotate is 90, 180 or 270. Returns the new artifact_id and a download_url (a link that works for about an hour). Page selections: https://getemboss.ai/docs/artifacts.
extract
extract_pages
Free. Keep only the pages you name from a PDF, in the order you name them. Keep only the pages named in pages (1-based, like 2,5-9, repeats and any order allowed) from artifact_id. Returns the new artifact_id and a download_url (a link that works for about an hour).
fax
get_fax
Delivery status and receipt of a send_fax job. status is queued, working, delivered, failed or expired; keep checking about every 20 seconds until it is delivered, failed or expired. A busy, unanswered or dropped call is retried automatically up to three more times over about 17 minutes while the status stays working; attempts, max_attempts and next_attempt_at say where it is. A failed or expired fax carries failure_category, failure_reason, retryable, refund and a plain failure_message saying what happened.
flatten
flatten_pdf
Free. Lock a filled form: draw every field's value and every comment into the page so nothing can be edited. Lock artifact_id, usually the filled PDF from fill_form: every field's value, checkbox and comment becomes part of the page, no fields stay editable, links keep working, and the original stays as it is. Returns the new artifact_id, a download_url (a link that works for about an hour) and a report: fields_locked, annotations_locked, pages, and calculated_left_blank (the labels of boxes the form totals itself that were empty). A locked form can't be read back or checked, so run read_form or verify_form first. Refuses a digitally signed PDF (pdf_signed) and a dynamic XFA form (xfa_dynamic).
forms
list_forms
Forms in the user's Emboss library. state: ready (default), processing, failed, or all. Paginate with cursor.
inspect
inspect_pdf
Free. Report a PDF's structure: page count, page sizes and rotations, encryption, form fields, annotations. Structural facts about artifact_id: page count, page sizes and rotations, whether it is encrypted, has form fields, annotations, outlines, embedded files or JavaScript. No text, no field values.
job
get_job
Status of a fill_form_from_context, prepare_form, or commit_proposal job. When ready, returns the download link, how many fields were filled, and for a commit, the result, receipt_url and, when one was asked for, package_url.
merge
merge_pdf
Free. Join whole PDFs into one file in the order given. Join whole PDFs in order: sources is a list of {artifact_id}. Returns the new artifact_id and a download_url (a link that works for about an hour). For page selections use compose_pdf.
pages
delete_pages
Free. Remove the pages you name from a PDF and keep the rest in order. Remove the pages named in pages (1-based, like 1-2,17) from artifact_id; every other page stays in order. Returns the new artifact_id and a download_url (a link that works for about an hour).
prepare
prepare_form
Prepare a form from documents: a proposal of every answer with its source, what is missing, and where documents disagree, before writing, plus attachments for a package. Same inputs as fill_form_from_context (form_id for a form already in the library, or pdf_url/pdf_base64 for a new one, plus context_text and/or context_urls), but nothing is written into the PDF: read the proposal, answer what is missing, then commit. policy is safe (write high and medium confidence) or strict (high only). Returns a job_id and a proposal_id; poll get_job about every 20 seconds. Billed as one context fill; the first 5 each month are free. Optional retention: ephemeral or account_default (default: the account's own setting); it applies only when this call creates the form (pdf_url, pdf_base64 or library), since an existing form keeps the setting it was created with. Ephemeral processing: source and output documents are deleted 60 to 70 minutes after the last activity on a form.
proposal
get_proposal
Prepare a form from documents: a proposal of every answer with its source, what is missing, and where documents disagree, before writing, plus attachments for a package. Read a proposal by its proposal_id: every field's state, its candidate values with evidence, and the questions still open. No charge to read it.
reorder
reorder_pages
Free. Put a PDF's pages in a new order you give, every page exactly once. Reorder artifact_id's pages: order lists every page number exactly once, in the new order. Returns the new artifact_id and a download_url (a link that works for about an hour).
rotate
rotate_pages
Free. Rotate the pages you name by 90, 180 or 270 degrees. Rotate pages of artifact_id: rotations is a list of {pages, degrees} with degrees 90, 180 or 270, added clockwise to each page's current rotation. Returns the new artifact_id and a download_url (a link that works for about an hour).
send
send_fax
Send a PDF to a fax number and track delivery, priced per transmitted page. Give a destination in E.164 form (for example +15025551212; US, Canadian and Mexican numbers only) and the PDF: an artifact_id from any earlier result, a job_id from fill_form_from_context or commit_proposal once get_job reports ready, a form_id for the fillable form, or one PDF by pdf_url or pdf_base64, or a sources list of artifact ids (each with an optional pages range such as 1-3,7) faxed together as one packet. Returns a job id at once; poll get_fax for delivery. Billed per page at delivery; a failed fax is not charged. Needs a payment method on the account; get_usage reports fax_ready.
suggest
suggest_mapping
Propose which spreadsheet columns fill which PDF form fields. Given a CSV (csv_text or csv_url), proposes which column fills which field. Only the header row matters. Billed as one context fill (5 free each month).
usage
get_usage
This month's usage and how many free operations remain, plus the billing page link. It also says whether a payment method is on file (payment_method_on_file) and whether the account can send a fax today (fax_ready). Faxes need a payment method: check fax_ready before send_fax.
verify
verify_form
Check a filled form: required fields, allowed options, consistent choices, text that fits, empty signature boxes; returns complete, review required or incomplete. Give a form_id and the filled PDF (pdf_url: public https link, or pdf_base64). Billed as one verify; the first 5 each month are free.

Endpoints

URLTransportStateLatencyChecked
https://api.getemboss.ai/mcp streamable-http answering 349 ms 3 min ago

Alternatives to Emboss

same job, measured the same way
DocsAutomator
by docsautomator

Make documents in your brand from your data and send them for acceptance or signature.

answering
D
Docmill
by docmill

Generate PDF, DOCX, and XLSX documents from JSON data and create reusable templates via Docmill

7 installs/wk local only
Sign PDF
by portant

Send a PDF, a Google Doc or Markdown for e-signature from chat, and get the signed PDF back.

answering
Google Docs MCP
by a1-x-tech

MCP server for the Google Docs API: read and edit documents, styles, tables, images, comments.

27 installs/wk local only
PDFluent
by pdfluent

Read, search and fill PDFs as markdown, text, tables, forms. Your documents stay on your machine.

361 installs/wk local only
Extend
by extend

Turn documents into structured data: parse, extract, classify, split, and fill PDF forms.

answering
Document Processing
by filegraph

Extract text from documents, manipulate PDFs, and perform OCR on images.

9 tools answering
Membase
by membase

Long-term memory for you and your AIs: search memories, read the profile, add facts and documents.

answering

Emboss — questions

Answers built from our own checks of this server.

What can Emboss do?
It exposes 28 tools, read directly from the server on our last check. Among them: add_attachment, commit_proposal, compose_pdf, create_form, delete_form, delete_pages and 22 more. The full list with descriptions is on this page — we take it from the server itself via tools/list, not from a README. How MCP servers expose tools in the first place →
What is Emboss mostly used for?
Its tools cluster around form and fill. That is what this server is built to work with — the grouping comes from the actual tool names, not from a category we assigned.
Is Emboss working right now?
We send a real MCP handshake every 15 minutes. Over the last 24 hours 91 of 91 checks got a reply (100.0%), average response time 339 ms. The bar chart above shows every period we have measured.
How do I connect Emboss?
Copy the ready config from this page. We generate it for Claude Code, Claude Desktop, Codex, Cursor and VS Code, each in the form that client actually reads. It is a remote server, so there is nothing to install — the client connects to the address.
Does Emboss need an API key?
No. Emboss completed a full MCP handshake with us as an anonymous client and listed its tools without asking for anything. All 28 of them are readable on this page. This is what we observed, not what the docs claim.
How fast is Emboss?
It answers our handshake in 339 ms on average, which is faster than 47% of all working MCP servers we measure. The comparison comes from our own checks across the whole registry, every 15 minutes.
Is Emboss open source?
The code is open source: it is published under the MIT licence, written in Shell and 0 stars on GitHub. The licence covers the code in the repository, not the hosted service at api.getemboss.ai, which runs on its operator's own terms.