Skip to content

doc.cheap MCP server

cheap.doc/mcpdoc.cheapstreamable-httpv0.3.11Scanned

Passport, ID and MRZ recognition via doc.cheap – Free: 100 documents every month, then $0.01 each

6 tools: 1 can delete or cancel things, 1 change things and 4 only read.

Tools: 1 delete, 0 irreversible, 1 write, 4 read.
  • Reads private data · 3
  • Reads text others write · 5
  • Can send data out · 0
  • No findings

Tools listed without sign-in. Remote posture

With Anlyon

1 tool here can delete or cancel things. Govern it.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the doc.cheap MCP server can do

The doc.cheap MCP server is published by doc.cheap, a domain the MCP Registry verified. It has 6 tools. It can delete or cancel things (1 tool: delete_scan) and create or change data (1 tool: scan_document). 4 tools only read.

3 tools read private data (check_balance, search_docs and get_scan); 5 tools bring in text other people can write (scan_document, check_balance, list_scans and 2 more).

None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.

Its remote endpoint lists its tools without sign-in, and its tools changed once since Anlyon Radar first saw them.

Last scanned Nov 18, 2026, version 0.3.11.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

6 tools

Deletes or cancels 1Claude Code: deny
  • delete_scanPermanently delete one stored scan now, before its retention window would end.Reads text others writeNo findings

    Why it is in this group

    • The name "delete_scan" says "delete": a word for removing or ending something.
    • The description says "Permanently delete".
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Permanently delete one stored scan now, before its retention window would end. This cannot be undone: the stored result, its history row and its thumbnail are removed, and the scan can no longer be listed, fetched or replayed through its idempotency_key. The credit it drew is not refunded, and this period's usage counters still count it. Input: scan_id – meta.id of a scan_document result, or id of a list_scans row. Output: { id, deleted: true }. Calls DELETE /v1/scans/{id}. Only scans made with a live key under a non-zero retention window are stored for the account, and only until that window ends (retain_hours on the scan, or the account's history-retention setting, one year by default); a scan made with retain_hours 0 was never stored. A key reaches its own account's scans and no other account's. Under a sandbox key nothing can be read back: an account's own sk_sandbox_ key stores nothing, and the public key keeps every scan for one year, with a reduced thumbnail and the address it came from, where that key cannot read, list or delete it. Errors: an id that is unknown, belongs to another account, has passed its window or was already deleted is refused as not_found, and nothing is deleted. Use it when someone asks for a document's data to be removed; confirm the id with list_scans or get_scan first, because the deletion is final.

    Annotations the server declares

    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: true
    • openWorldHint: true
Changes things 1Claude Code: ask
  • scan_documentRecognise a passport, national ID card or driver's licence from a photo or scan and return what is printed on it as structured JSON.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Recognise a passport, national ID card or driver's licence from a photo or scan and return what is printed on it as structured JSON. Inputs: the image as image_base64 or image_url (https, on a public address); plus the optional expect_country, return_portrait, retain_hours, reference and idempotency_key. Output: a Scan object – meta (id, status, billed, confidence, timing), document (kind, issuing country, number, series, date of issue, date of expiry, whether it has expired and how many days are left), holder (given names, surname, date of birth, sex, nationality) – country codes are the document's own three-letter ICAO codes, each with a *_detail object giving the code as written, its ISO 3166-1 alpha-3 and alpha-2 and its category, fields (every field read off the printed page, each with its own confidence), mrz (whether the machine-readable zone checks out, why not when it does not, and its lines exactly as read), images, quality and authenticity – plus a one-line summary of the same result. Calls POST /v1/scans. Cost: it bills one credit ($0.01) only when a document is recognised; an unreadable image, an empty frame or an unsupported type costs nothing, and meta.billed says which happened. Without a key, the public sandbox key is used. It gives 10 free recognised documents per address in all, and at most 10 requests per address an hour, whatever their answer. Registering gives 100 free documents every month. Use test images only, never a real document. Use it whenever someone hands over an identity document and wants it read, transcribed, or checked against what they claim – a name, a document number, a date of birth or an expiry date.

    Annotations the server declares

    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true
Only reads 4Show 4 read-only tools
  • check_balanceReturn how many credits are left on the account and what the current period has used: the balance, split into this month's free credits (100 every month, drawn first) and the paid …Reads private dataReads text others writeNo findings

    Why it is in this group

    • The description's first verb is "return": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "return" and "document": it reads documents.
    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Return how many credits are left on the account and what the current period has used: the balance, split into this month's free credits (100 every month, drawn first) and the paid credits, the credits spent, and the scan counters broken down by status (recognized, unreadable, no document found, unsupported document, rejected). Takes no arguments and calls GET /v1/usage; under the public sandbox key it answers without calling anything. One recognised document draws one credit, at $0.01; scans that recognised nothing are counted and never charged. Needs a real API key – under the public sandbox key there is no account behind the call, and the answer says so instead of reporting zeros that read like a balance. Use it before working through a batch of documents, or when a scan is refused for lack of credit.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
  • get_scanFetch the full result of one stored scan by its id, as it was returned when the document was recognised.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_scan" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "fetch" and "document": it reads documents.
    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Fetch the full result of one stored scan by its id, as it was returned when the document was recognised. Input: scan_id – meta.id of a scan_document result, or id of a list_scans row. Output: the same Scan object scan_document returns (meta, document, holder, fields, mrz and the rest) plus a one-line summary, with two differences: the image crops are never stored, so every images slot is null, and quality reads not_checked. Calls GET /v1/scans/{id}; it never re-runs recognition and never charges a credit. Only scans made with a live key under a non-zero retention window are stored for the account, and only until that window ends (retain_hours on the scan, or the account's history-retention setting, one year by default); a scan made with retain_hours 0 was never stored. A key reaches its own account's scans and no other account's. Under a sandbox key nothing can be read back: an account's own sk_sandbox_ key stores nothing, and the public key keeps every scan for one year, with a reduced thumbnail and the address it came from, where that key cannot read, list or delete it. Errors: an id that is unknown, belongs to another account, was made with retain_hours 0 or has passed its window is refused as not_found – the cases are not told apart. Use it to read back a document recognised earlier instead of scanning the image again.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
  • list_scansList the account's stored scans, most recent first, one page at a time.Reads text others writeNo findings

    Why it is in this group

    • The name "list_scans" says "list": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    List the account's stored scans, most recent first, one page at a time. Inputs: the optional limit (1 to 100 rows, default 20) and cursor (the next_cursor of the previous page; omit it for the first page). Output: scans – one row per scan with id, status (recognized, unreadable, no_document_found, unsupported_document or rejected), billed, duration_ms, reference (your own string from the scan) and created_at – and next_cursor, which is null on the last page. A row holds no extracted data; call get_scan with its id for the full result. Calls GET /v1/scans; it never charges a credit. Only scans made with a live key under a non-zero retention window are stored for the account, and only until that window ends (retain_hours on the scan, or the account's history-retention setting, one year by default); a scan made with retain_hours 0 was never stored. A key reaches its own account's scans and no other account's. Under a sandbox key nothing can be read back: an account's own sk_sandbox_ key stores nothing, and the public key keeps every scan for one year, with a reduced thumbnail and the address it came from, where that key cannot read, list or delete it. Under the public sandbox key the answer is an empty list, given without calling the API. Errors: a cursor this API did not issue is refused (validation_failed); a key the API does not know is refused (unauthorized). Use it to find a scan made earlier – by its reference or its time – before reading or deleting it.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
  • search_docsFull-text search over the doc.cheap API documentation – endpoints, request options, every response field, the error codes and what to do about each, MRZ rules, retention and pricin…Reads private dataNo findings

    Why it is in this group

    • The name "search_docs" says "search": a word for reading.

    What it can reach

    • private-dataThe name "search_docs" says "search" and "doc": it reads documents.
    • private-dataThe description's first sentence says "doc": it reads documents.
    Description, as the server wrote it
    Full-text search over the doc.cheap API documentation – endpoints, request options, every response field, the error codes and what to do about each, MRZ rules, retention and pricing. Takes a query and an optional limit (1 to 20, default 5), and answers with the matching sections: title, a snippet, and a link to the page. It reads a copy of the documentation shipped beside this server, so it makes no network call and works offline. Use it before guessing at a field name, an error code or a scan option – what it returns is the published contract rather than a recollection of it.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: false

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

No private-data, untrusted-input, outbound path in this server alone: no tool is tagged outbound.

Reads private data3
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write5
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out0
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 18:31 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
2
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed
  • HighThe whole server

    The endpoint served tools/list without credentials, and the list has 2 tools that change things (1 destructive, 1 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://mcp.doc.cheap/mcp: tools/list returned 6 tools to an anonymous client, including delete_scan (destructive), scan_document (write)

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

No check matched anything in these tool definitions. That is not a safety guarantee.

History

Changes to its tool definitions and to its remote endpoint, newest first.

  1. Tools4 changed.~4 changedView diff

Install

Pinned to version 0.3.11. Values in angle brackets are yours to fill in.

claude mcp add --transport http doc https://mcp.doc.cheap/mcp
claude mcp add --transport http doc https://mcp.doc.cheap/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__doc__check_balance",
      "mcp__doc__search_docs",
      "mcp__doc__list_scans",
      "mcp__doc__get_scan"
    ],
    "ask": [
      "mcp__doc__scan_document"
    ],
    "deny": [
      "mcp__doc__delete_scan"
    ]
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as doc. Review it before you use it.

Capability tags behind 6 rules
  • mcp__doc__scan_document untrusted-input
  • mcp__doc__check_balance private-data, untrusted-input
  • mcp__doc__search_docs private-data
  • mcp__doc__list_scans untrusted-input
  • mcp__doc__get_scan private-data, untrusted-input
  • mcp__doc__delete_scan untrusted-input

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/cheap.doc/mcp/verify?hash=9339c4bb83c0cf52c140484edfbdb6a79672b17e6b1ec570d511a40d598fce39" | jq -e '.changed == false'

Pinned hash 9339c4bb83c0, from Nov 18, 2026, 11:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/cheap.doc/mcp/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for cheap.doc/mcp
[![Anlyon Radar](https://radar.anlyon.com/badge/cheap.doc/mcp.svg)](https://radar.anlyon.com/servers/cheap.doc/mcp)

Source: MCP Registry · repository. Synced Nov 4, 2026, 18:17 UTC.

doc.cheap MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the doc.cheap MCP server safe?

No scan can promise that. What its definitions show: 6 tools, 2 of which can change, send or delete something, and 0 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the doc.cheap MCP server delete data?

Yes. 1 tool can delete, cancel or revoke something: delete_scan.

Can the doc.cheap MCP server do things that cannot be undone?

No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.

What tools does the doc.cheap MCP server have?

6 tools: 1 destructive, 1 write, 4 read. delete_scan, scan_document, check_balance, search_docs, list_scans and get_scan.

Does the doc.cheap MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the doc.cheap MCP server send data to outside services?

None of its tools sends, posts or publishes, or takes a recipient, channel or URL.

Has the doc.cheap MCP server changed its tools?

Yes, once since Anlyon Radar started tracking it, most recently on Nov 4, 2026. Every change is on this page and in its Atom feed.