Name: mint utilities
Author: voiddo

mint utilities MCP

26 deterministic tools: PDF, image, QR, CSV/Excel, HTML to PDF, DNS/RDAP, FX rates, time zones.

Tool-list grade A Remote · Streamable HTTP No auth needed 26 tools v1.1.2

No reviews yet by voiddo The publisher’s own endpoint — this marketplace is not in the path

93/100

tool-list score

tools declared

remote

at its publisher

—

calls last 7d

What it does

26 deterministic tools: PDF, image, QR, CSV/Excel, HTML to PDF, DNS/RDAP, FX rates, time zones.

Quickstart

# 1 — add it to Claude Code
claude mcp add --scope user --transport http mint-utilities-mcp https://data.voiddo.com/mcp/

# 2 — ask your agent for something these tools do
#   csv_profile, currency_rates, dns_lookup and 23 more

mint utilities costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.

Publishernot claimed

voiddo1 server

1 category

All 1 server

Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.

Tool surface

What the model actually sees. Descriptions are diffed on every release — see version history.

Only tools that write or reach the network

Profile a CSV, TSV, XLSX or JSON table: row and column counts plus type, nulls, distinct values, min, max and samples for every column. Cost: 1 credit per call. File input (file): base64 string, data: URI or {"url": "https://..."}. Output: JSON. {rows, cols, columns: [{name, inferred_type, null_count, distinct_count, distinct_capped, min, max, sample}]}

Parameter	Type	Required	Notes
file	string/object	yes	The table: base64 string (or data: URI) or {"url": "https://..."}. Up to 10 MB.
from	string	no	Source format. Omit to auto-detect.
sheet	string/integer	no	xlsx input: sheet name, or 0-based sheet index (default: the first sheet).
header	boolean	no	The first row holds column names. false: columns are named col1, col2, ...
delimiter	string	no	csv input: field delimiter (default: detected among , ; tab \|).
encoding	string	no	Text encoding of csv/tsv/json input (utf-8 falls back to cp1252).
max_rows	integer	no	Reject inputs with more data rows than this (413).

Latest and historical euro foreign exchange reference rates from the European Central Bank, with conversion between any two of its currencies. Cost: 1 credit per call. Output: JSON. latest: {base, date, rates}; convert: {amount, from, to, rate, result, date_used}; history: {base, quote, series[{date, rate}]}

Parameter	Type	Required	Notes
op	string	no	—
base	string	no	Base currency (latest, history); for convert the alias of 'from'
quote	string	no	Quote currency (history; latest filter; convert alias of 'to')
quotes	array	no	Several quote currencies for latest; omit for all
amount	number	no	Amount of the 'from' currency (convert)
from	string	no	Currency to convert from (default EUR)
to	string	no	Currency to convert to
date	string	no	YYYY-MM-DD; non-business days use the previous business day. Default: latest day
days	integer	no	history: calendar days ending at 'date' (default the latest day)

Look up A, AAAA, MX, TXT, NS, CNAME, SOA, CAA, SRV or PTR records of a public name through the Cloudflare and Google public resolvers. Cost: 1 credit per call. Output: JSON. {name, type, status, records:[{type, value, ttl}], count, resolver, elapsed_ms}

Parameter	Type	Required	Notes
name	string	yes	Public hostname, e.g. example.com (internationalised names are converted to punycode). For type PTR: an IPv4/IPv6 address or a complete in-addr.arpa / ip6.arpa name.
type	string	no	Record type
all	boolean	no	Query the common set (A, AAAA, CNAME, MX, NS, TXT, SOA, CAA) in one call; 'type' is then ignored

Public holidays of any of about 250 countries and regions for a given year, with optional subdivision, language and category. Cost: 1 credit per call. Output: JSON. {country, year, subdiv, holidays:[{date, name}], count}

Parameter	Type	Required	Notes
country	string	yes	ISO 3166-1 alpha-2 country code, e.g. US, DE, IL
year	integer	yes	—
subdiv	string	no	State / province / region code or name, e.g. BY, Bayern, CA, SCT
language	string	no	Language of the holiday names, e.g. de or en_US; default: English where available, else the country's own language
categories	array	no	Holiday categories, default ["public"]; e.g. public, bank, school, government, optional, half_day
observed	boolean	no	Include substitute days such as "Christmas Day (observed)"

Render an HTML snippet to a PNG, JPEG or WEBP image at any width, with retina scaling and transparent background. Cost: 3 credits per call. Output: a file (image/png | image/jpeg | image/webp) returned inline, at most 1 MiB. The rendered image

Parameter	Type	Required	Notes
html	string	yes	The HTML document or fragment (UTF-8, up to 2 MB). JavaScript is not executed.
width	integer	no	Viewport width in CSS px
height	integer	no	Viewport height in CSS px; omit for a full-page image that fits the content
format	string	no	—
quality	integer	no	jpeg / webp quality
device_scale_factor	number	no	Pixel density (2 = retina)
transparent	boolean	no	Transparent page background (png / webp only)

Turn an HTML document into a print-ready PDF with page size, margins, orientation and page-numbered header and footer. Cost: 3 credits per call. Output: a file (application/pdf) returned inline, at most 1 MiB. The PDF document

Parameter	Type	Required	Notes
html	string	yes	The HTML document or fragment (UTF-8, up to 2 MB). JavaScript is not executed.
format	string	no	Paper size
landscape	boolean	no	—
margin_mm	number	no	Page margin on all sides, millimetres
print_background	boolean	no	Print CSS backgrounds and colours
scale	number	no	Rendering scale
header_text	string	no	Plain-text page header; {page} and {pages} are replaced
footer_text	string	no	Plain-text page footer; {page} and {pages} are replaced

Convert and resize images between PNG, JPEG, WEBP, GIF, BMP, TIFF and ICO, with quality, fit modes, transparency handling and optional metadata stripping. Cost: 1 credit per call. File input (file): base64 string, data: URI or {"url": "https://..."}. Output: a file (image/png | image/jpeg | image/webp | image/gif | image/bmp | image/tiff | image/x-icon) returned inline, at most 1 MiB. The converte

Parameter	Type	Required	Notes
file	string/object	yes	The image: base64 (or data: URI) string, or {"url": "https://..."}; up to 10 MB
to	string	yes	Target format
width	integer	no	Target width in px
height	integer	no	Target height in px
fit	string	no	—
quality	integer	no	jpeg / webp quality (webp 100 = lossless)
strip_metadata	boolean	no	Drop EXIF / XMP / ICC (colours are converted to sRGB first)
background	string	no	Colour for transparent areas when the target has no alpha, and for contain padding (default white / transparent)

Reverse DNS name and registry record (network block, name, country) of a public IP address, from open registry data only. Cost: 1 credit per call. Output: JSON. {ip, version, is_private, reverse_dns, reverse_dns_all, network{name, country, ...}}

Parameter	Type	Required	Notes
ip	string	yes	Public IPv4 or IPv6 address

Look up ISO 3166 country and subdivision codes, ISO 4217 currency codes and ISO 639 language codes by code or name fragment. Cost: 1 credit per call. Output: JSON. {type, query, results:[{alpha_2, alpha_3, numeric, name, ...}], count, total, truncated}

Parameter	Type	Required	Notes
type	string	yes	—
query	string	yes	Code or name fragment, case- and accent-insensitive
limit	integer	no	—
country	string	no	Only for type=subdivision: restrict to this country (ISO 3166-1 alpha-2)

Validate a JSON document, or up to 1000 of them, against a JSON Schema (draft 7, 2019-09 or 2020-12) and list every error with its path. Cost: 1 credit per call. Output: JSON. {valid, errors: [{index, path, message, validator}], count}; meta: {draft, instances, truncated?}

Parameter	Type	Required	Notes
schema	object	yes	The JSON Schema to validate against.
instance		no	One JSON value to validate (any type, including null). Use this or instances.
instances	array	no	Up to 1000 JSON values validated against the same schema. Use this or instance.
draft		no	JSON Schema draft. auto follows the schema's $schema and defaults to 2020-12.

Convert text between JSON and YAML with safe parsing, indent, key sorting and minify options, and line-accurate error messages. Cost: 1 credit per call. Output: a file (application/json | application/yaml) returned inline, at most 1 MiB. converted.json or converted.yaml; meta: {from, to}

Parameter	Type	Required	Notes
input	string	yes	The JSON or YAML text to convert (up to 2 MB).
from	string	no	Source format; auto tries JSON first, then YAML.
to	string	no	Target format. Default: yaml for JSON input, json for YAML input.
indent	integer	no	Spaces per level. JSON: 0 puts items on their own lines without indentation. YAML: values below 2 are treated as 2.
sort_keys	boolean	no	Sort object keys alphabetically (default: keep input order).
minify	boolean	no	JSON: single line without spaces. YAML: single-line flow style. Overrides indent.

Check a KDP paperback interior and cover PDF against Amazon's published file specs and get a signed report. Cost: 5 credits per call. File inputs (interior, cover): base64 string, data: URI or {"url": "https://..."}. Output: JSON. {id, verdict: pass|warn|fail, label, ruleset, specs_date, created_at, files: {interior, cover?: {name, sha256, bytes}}, interior: {trim, pages, bleed, gutter_in, ...}, c

Parameter	Type	Required	Notes
interior	string/object	yes	The interior PDF: base64 string (or data: URI) or {"url": "https://..."}. Up to 10 MB.
cover	string/object	no	Optional wrap cover PDF (back, spine and front in one page, with bleed).
trim	string	no	Trim size; found from the page size when omitted
paper	string	no	—

A ready-to-upload KDP paperback puzzle book: interior PDF, wrap cover, signed file check, previews and listing notes. Cost: 500 credits per call. Output: a file (application/zip) returned inline, at most 1 MiB. ZIP: interior.pdf, cover.pdf, check-report.pdf/.json, preview/*.png, listing.txt, README.txt

Parameter	Type	Required	Notes
kind	string	yes	sudoku, wordsearch, maze, cryptogram, wordscramble, numbersearch, kakuro, fillin, killersudoku, calcudoku, futoshiki, nonogram, skyscrapers, logicgrid, crossword, numberfill, samuraisudoku, jigsawsudo
count	integer	yes	Number of puzzles. The fewest allowed fill KDP's 24-page minimum; the most are 400 sudoku (300 at hard or expert), 150 word searches, 200 mazes, 300 cryptograms, 55 word scrambles (40 at easy or hard)
level	string	yes	easy, medium, hard (and expert for sudoku and mazes). For kidssudoku the level picks the grid: easy 4x4, medium 6x6, hard 9x9, mixed all three in that order. For kidswordsearch it picks the grid and t
trim	string	no	—
paper	string	no	—
large_print	boolean	no	Every text read while solving at 16 pt or more. Nonograms need room: in large print, easy on 7x10 or larger and medium on 8.5x11 or 8.27x11.69 (no hard); in regular print, hard on 7x10 or larger. Logi
title	string	no	—
subtitle	string	no	—
author	string	no	—
themes	array	no	Word search, word scramble, fill in or crossword themes to use, in order (default: the everyday themes, shuffled; the Christmas and winter ones only by name)
mix	array	no	Variety books only: the kinds of puzzle, in book order (default: wordsearch, sudoku, cryptogram, maze). Kids kinds only (kidssudoku, kidswordsearch, kidsmaze) make an activity book for kids; they do n
cover_color	string	no	The cover's color, so a series can look like no one else's. Default: the kind's own (sudoku navy, word search forest, maze terracotta, ...).
cover_layout	string	no	Where the title sits on the front cover: puzzle_top (default: a puzzle from the book above the title), title_top (the title above the puzzle) or title_card (the puzzle across the whole front, the titl
seed	integer	no	Rebuild an earlier book exactly; omit for a new one

Look up a KDP file check report by id and confirm its signature and file fingerprints. Free tier only (credits do not apply to this tool). Output: JSON. {found, genuine, report?}

Parameter	Type	Required	Notes
id	string	yes	Report id (20 hex characters)

Full cover size, spine width, safe areas, barcode box, gutter and margins for an Amazon KDP paperback, with an optional PDF cover guide. Cost: 1 credit per call. Output: JSON. {valid, problems[], ruleset, specs_date, trim, pages, paper, page_limits, cover: {width_in, height_in, spine_in, bleed_in, width_px_300dpi, height_px_300dpi, spine_text_allowed, back_safe, front_safe, spine_text_box, barcode

Parameter	Type	Required	Notes
trim	string	no	Trim size in inches, width x height
pages	integer	yes	Interior page count
paper	string	no	white, cream, groundwood, standard_color or premium_color
bleed	boolean	no	Interior pages with bleed (pictures to the edge)
template	boolean	no	Also return a PDF cover guide (base64)

Render Markdown to a sanitised HTML page or fragment, or to a PDF, with tables, fenced code, heading anchors, an optional table of contents and a choice of themes. Cost: 1 credit per call. Output: a file (text/html | application/pdf) returned inline, at most 1 MiB. document.html / fragment.html (text/html; charset=utf-8) or document.pdf. meta: {format, theme, title, headings, bytes, pages?}. pdf c

Parameter	Type	Required	Notes
markdown	string	yes	The Markdown source (up to 1 MB of text)
format	string	no	html = complete page (1 credit), fragment = content only (1 credit), pdf = A4 document (3 credits)
title	string	no	Page title; default is the first level-1 heading
theme	string	no	Built-in stylesheet for html and pdf
css	string	no	Alias of theme
toc	boolean	no	Add a table of contents above the content

Read a PDF's page count, encryption flag, metadata, page sizes in millimetres, text-layer presence and bookmark count. Cost: 1 credit per call. File input (file): base64 string, data: URI or {"url": "https://..."}. Output: JSON. {page_count, encrypted, password_required, metadata: {title, author, subject, creator, producer}, page_sizes_mm: [{n, width, height}] (first 50 pages), has_text, outline_c

Parameter	Type	Required	Notes
file	string/object	yes	The PDF: base64 string (or data: URI) or {"url": "https://..."}. Up to 10 MB.

Merge 2 to 20 PDF files into one document in the order given, optionally keeping their bookmarks. Cost in credits: ceil(total pages / 20), minimum 1. File input (files): base64 string, data: URI or {"url": "https://..."}. Output: a file (application/pdf) returned inline, at most 1 MiB. merged.pdf; meta: {page_count, files}

Parameter	Type	Required	Notes
files	array	yes	2 to 20 PDFs in merge order; all together up to 10 MB.
keep_bookmarks	boolean	no	Keep the bookmarks (outline) of the input files.

Split a PDF into parts by page ranges such as '1-2,5-' or every N pages, returned as one PDF or a zip of parts. Cost in credits: ceil(pages written / 20), minimum 1. File input (file): base64 string, data: URI or {"url": "https://..."}. Output: a file (application/pdf | application/zip) returned inline, at most 1 MiB. part_01.pdf when there is one part, otherwise split.zip holding part_01.pdf ...;

Parameter	Type	Required	Notes
file	string/object	yes	The PDF: base64 string (or data: URI) or {"url": "https://..."}. Up to 10 MB.
ranges	string	no	Comma-separated page ranges, one part each: '1-2,5-' ('5-' runs to the last page).
every_n	integer	no	Alternative to ranges: parts of this many pages.

Extract the text of a PDF page by page as plain text or Markdown, with page selection, an optional layout-preserving mode and the document metadata. Cost in credits: ceil(pages extracted / 20), minimum 1. File input (file): base64 string, data: URI or {"url": "https://..."}. Output: JSON. {page_count, pages: [{n, text}] (Markdown: [{n, markdown}]), chars, metadata: {title, author, subject, keyword

Parameter	Type	Required	Notes
file	string/object	yes	The PDF: base64 string (or data: URI) or {"url": "https://..."}. Up to 10 MB.
pages	string	no	Pages to extract, e.g. '1-3,7' or '5-' (default: all pages).
layout	boolean	no	Preserve the visual layout (columns, spacing) instead of reading order.
format	string	no	text: the plain text of each page. markdown: headings, bold and italic, lists and paragraphs as Markdown, page headers and footers that repeat on most pages left out (layout does not apply).

Turn text or a URL into a crisp QR code as PNG or SVG, with size and error-correction control. Cost: 1 credit per call. Output: a file (image/png | image/svg+xml) returned inline, at most 1 MiB. The QR image

Parameter	Type	Required	Notes
text	string	yes	Text or URL to encode
format	string	no	—
size	integer	no	Edge length in px (png)
margin	integer	no	Quiet zone in modules
error_correction	string	no	—
fg	string	no	—
bg	string	no	—

Look up the public registration record of a domain, IP address or AS number from its registry's RDAP service, without personal contact data. Cost: 1 credit per call. Output: JSON. Normalised registration record: domain, IP network or autonomous system

Parameter	Type	Required	Notes
query	string	yes	Domain name (example.com), IPv4/IPv6 address (1.1.1.1) or AS number (AS15169)
fields	string	no	compact: normalised summary; full: summary plus the registry answer without contact data under 'raw'

Capture a public web page as a PNG, JPEG or WEBP image: any viewport size, full page, retina scaling, optional cookie-banner hiding. Free tier only (credits do not apply to this tool). Output: a file (image/png | image/jpeg | image/webp) returned inline, at most 1 MiB. The screenshot

Parameter	Type	Required	Notes
url	string	yes	Public http(s) URL (robots.txt is honoured), or an inline data:text/html URL
width	integer	no	Viewport width in CSS px
height	integer	no	Viewport height in CSS px
full_page	boolean	no	Capture the whole scrollable page (capped at 16384 CSS px / 50 megapixels)
format	string	no	—
quality	integer	no	jpeg / webp quality
delay_ms	integer	no	Extra wait after load, before the shot
device_scale_factor	number	no	Pixel density (2 = retina)
block_cookie_banners	boolean	no	Hide common cookie-consent dialogs (best effort)

Render an SVG document to a PNG at a chosen width, height or scale and background; external files, network references and XML entities are never loaded. Cost: 1 credit per call. File input (file): base64 string, data: URI or {"url": "https://..."}. Output: a file (image/png) returned inline, at most 1 MiB. The rendered PNG

Parameter	Type	Required	Notes
svg	string	no	The SVG markup as text (up to 2 MB). Use file instead for a base64 / URL / uploaded .svg or .svgz
file	string/object	no	Alternative to svg: an .svg or .svgz file as base64, data: URI or {"url": "https://..."}
width	integer	no	Output width in px (height follows the aspect ratio unless given)
height	integer	no	Output height in px (width follows the aspect ratio unless given)
scale	number	no	Multiplier of the SVG's own size (default 1); not combinable with width / height
background	string	no	transparent, #rrggbb or #rrggbbaa

Convert tables between CSV, TSV, XLSX and JSON with format, delimiter and encoding detection, header control and exact number and date handling. Cost in credits: 1 credit; 2 credits when the output is xlsx. File input (file): base64 string, data: URI or {"url": "https://..."}. Output: a file (text/csv | application/vnd.openxmlformats-officedocument.spreadsheetml.sheet | application/json) returned

Parameter	Type	Required	Notes
file	string/object	yes	The table: base64 string (or data: URI) or {"url": "https://..."}. Up to 10 MB.
from	string	no	Source format. Omit to auto-detect.
to	string	yes	Target format.
sheet	string/integer	no	xlsx input: sheet name, or 0-based sheet index (default: the first sheet).
header	boolean	no	The first row is a header (becomes JSON keys). false: no header row is read or written, and JSON output is an array of arrays.
delimiter	string	no	csv input: field delimiter (default: detected among , ; tab \|).
encoding	string	no	Text encoding of csv/tsv/json input. utf-8 falls back to cp1252; any other value is strict.
max_rows	integer	no	Reject inputs with more data rows than this (413) instead of truncating silently.
infer_types	boolean	no	csv/tsv input to json/xlsx output: turn exact numbers (and, for xlsx, ISO dates) and true/false into typed values.

Convert a date and time between IANA time zones with exact daylight-saving handling, inspect a zone's offset and transitions, or list a country's zones. Cost: 1 credit per call. Output: JSON. convert: {input, output, offset_from, offset_to, dst_from, dst_to, utc, ...}; info: {tz, offset, abbreviation, dst, previous_transition, next_transition}; list: {country, zones, count}

Parameter	Type	Required	Notes
op	string	no	—
datetime	string	no	ISO 8601, e.g. 2026-03-29T01:30:00 (wall clock in from_tz / tz) or 2026-03-29T00:30:00Z / 2026-03-29T01:30:00+01:00 (exact instant). Required for convert and info.
from_tz	string	no	convert: source zone, an IANA name such as Europe/Berlin or a fixed offset such as +05:30. Required when datetime has no offset.
to_tz	string	no	convert: target zone, an IANA name or a fixed offset. Required for convert.
tz	string	no	info: the zone to describe, an IANA name or a fixed offset (from_tz is accepted as an alias). Required for info.
country	string	no	list: ISO 3166-1 alpha-2 code, e.g. US. Omit to list every zone.
query	string	no	list: only zones whose name contains this text (case-insensitive), e.g. york or europe/

Free

$0 forever

Every tool — any call limit is its publisher’s, not ours
No card, and mcprush charges nothing for it
Runs at its publisher’s own address — this marketplace is not in the path
Nothing to cap, because mcprush bills nothing

Nothing here is billable: mcprush charges nothing to install mint utilities or to call it, at any volume.

ATool-list grade · score 93 / 100

Release 1.1.2 · scanned 29 Oct 2026

Graded from the tools this server lists, not from its source code.

Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.

Threat score100

Client score93

Findings0 threat · 1 capability

Capabilitymoderate

Coveragemanifest

Verificationnot in this report

Surface digestd0aef5a6a5b3fff7cab41032fa9ec45c8747632b8cbacd2a46a05498c15798c6

Score vectorthreat 100 → client 93

Term	Level	What it prices	Points
capability-exposure	moderate	capability blast radius (moderate) — client exposure if the model is manipulated	−3
coverage-honesty	manifest	inspection depth (manifest) — how much of the target the scan could see	−4

score = clamp(round(threatScore − E_cap − E_ver − E_cov), 0, 100). No model, no opinion, no network call at scoring time: the same methodology version against the same bytes always produces this number, and every line above is a field of the report rather than a summary of it.

What the scan could actually read

A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.

The scan read each tool’s name, description, input schema and annotations exactly as the server listed them the last time it answered us; no code was reviewed. We ask the server again about once a month, and a list that changed is graded again.

Tool surfaceread

Implementation sourcenot available

Package metadatanot available

Live transportnot available

The implementation source was not read — this reflects the declared tool metadata only.

Capability — what it could do if the model were manipulated

Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. moderate is the level these add up to.

Tool	Capability tags	Why the tag was assigned
csv_profile	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
currency_rates	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
dns_lookup	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
holidays_lookup	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
html_to_image	no tags
html_to_pdf	no tags
image_convert	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
ip_lookup	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
iso_codes	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
json_schema_validate	no tags
json_yaml_convert	no tags
kdp_preflight	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
kdp_puzzle_book	no tags
kdp_report	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
kdp_spine_calc	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
markdown_render	no tags
pdf_info	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
pdf_merge	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
pdf_split	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
pdf_text	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
qr_code	no tags
rdap_lookup	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
screenshot	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
svg_to_png	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
tabular_convert	untrusted-input	untrusted-input: declares openWorldHint: true (interacts with entities outside the server)
timezone_convert	no tags

Toxic-flow graph

The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.

The public result for this release does not print the flow graph, so there is nothing to show here. That is not the same as "no paths were found": what the scan did read is above, under coverage.

Supply chain and provenance

Integrityfirst-seen

Surface digestd0aef5a6a5b3fff7

Scanned2026-10-29

This is the first scan of this surface here, so there is nothing yet to compare it against.

Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.

Where are you running it?

Tool surface

Release history

mint utilities MCP

What it does

Quickstart

Where are you running it?

Tool surface

What counts against your monthly calls

What the scan could actually read

Capability — what it could do if the model were manipulated

Toxic-flow graph

Supply chain and provenance

Release history

See also