QAmcp

Copied. Paste it into Claude Code, Cursor or Codex:Fetch and follow the instructions at https://qamcp.dev/setup.md to test my MCP server.

Your AI finds your server in your project, tests it with QAmcp, explains what's wrong in plain words, fixes it with your OK, and tests again. Read the instructions

NewYour score can change any day. QAmcp Monitor keeps testing your server and emails you when it breaks. Free for 1 server.See what Monitor does →

Directory / MyDataHub

MyDataHub

Q: Is MyDataHub MCP server safe?

QAmcp's test on 9 Nov 2026: None of the hidden-instruction wordings we know were found in its 34 tool descriptions. Its error messages showed no secrets or internal details. 25 of its 34 tools are labeled read-only; the others may change or send things. Overall it scored 89/100 on 9 Nov 2026. A test can only see what the server shows from outside, not what it does with your data behind the scenes. Connect servers from makers you trust.

by emailmovers.com · https://mydatahub.emailmovers.com/mcp

UK B2B contact counts, quotes and lists, by Emailmovers

34 tools✓ no account neededPricing: not listed

Visit website

89 /100

PASS · WITH NOTES

12 of 12 checks · 9 Nov 2026

Test it now for current results

At a glance

Runs where: On their server. What you send to it goes to them.
What you need: Nothing: it answered without a login
Tools: 34 tools
Its own safety labels: 25 read-only, 9 labeled as changing things
Memory it takes: About 21,000 tokens of your AI's memory, just for its tool descriptions
Largest answer: About 16.2 KB in our test
Tools that look alike: 33 pairs of tools with nearly the same description, so the AI may pick the wrong one
Speed: Average answer in our test: 217 ms

Tokens are an estimate: the characters in its tool names, descriptions and inputs, divided by 4. The exact number depends on your AI model. Safety labels are what the server says about itself. From our latest test, 9 Nov 2026.

Is MyDataHub MCP server safe?

None of the hidden-instruction wordings we know were found in its 34 tool descriptions.
Its error messages showed no secrets or internal details.
25 of its 34 tools are labeled read-only; the others may change or send things.
Overall it scored 89/100 on 9 Nov 2026.

A test can only see what the server shows from outside, not what it does with your data behind the scenes. Connect servers from makers you trust.

What this test does not check

Its source code and packages.
What a tool quietly does behind the scenes, such as writing a file or calling another service.
Instructions hidden inside live tool answers. We only look for those in tool descriptions.
Hidden instructions worded in a way our list doesn't know yet.
Tools that could change something. Those are read, never run.

What QAmcp can and can't see · How to tell if an MCP server is safe · Hidden instructions · Prompt injection

Known security advisories

It lists no public code, so there is nothing to look up in GitHub Security Advisories or OSV.dev. QAmcp tests how a server behaves. It does not scan source code.

What it can do

34 tools, in the server's own words.

get_plansread-only

Returns the plans of MyDataHub by Emailmovers, the UK B2B count engine: credit packs, credit rules, licence terms, the bounce warranty and the free trial, in G…

get_accountread-only

Returns the organisation's credit balance by source, plan, per-action prices and rate limit, and for an agent key its monthly budget and per-order cap.

list_filtersread-only

Lists every filterable field of the audience definition (mydatahub.audience/1) with its group, type, operators, value count and default, plus the datasets.

search_filter_valuesread-only

Searches the values of one field by text, such as towns matching leeds, and returns each code with its contact and company counts plus did_you_mean for near mi…

expand_job_titlesread-only

Expands title keywords into every matching title (plurals, department synonyms, ampersands, abbreviations such as CEO), each with a count, or shows what a bund…

resolve_locationread-only

Turns free text such as 20 miles of Leeds, Yorkshire or LS1 into region, county, town, postcode or radius filters, each with a count and a ready filter.

compile_audienceread-only

Compiles a plain English brief, such as 300 finance directors at law firms in Manchester, into an audience definition with a one-sentence readback, editable ch…

count_audienceread-only

Counts UK contacts and companies matching an audience, inline or a saved aud_ id, and returns Total, New to you and Emailable, up to three breakdowns, quality,…

preview_recordsread-only

Returns up to 25 masked sample rows (10 by default) for an audience or a count id, each with verification badges, why it matches and a provenance card.

save_audience

Saves an audience definition under a name, shared across the organisation, or updates a saved one when audience_id is given.

list_audiencesread-only

Lists the organisation's saved audiences with their current counts and last refresh, newest first, optionally filtered by name.

list_field_setsread-only

Lists the field sets an order can deliver, such as standard or standard_telephone, with their exact columns.

quote_orderread-only

Creates a machine quote for a count: the records you would receive after removing ones already held and suppressed, the credits and GBP price, the licence, the…

create_order

Spends 1 credit per complete UK record against a live machine quote from quote_order, and queues the order.

get_orderread-only

Returns one order's status (queued, building, verifying, ready, failed or cancelled), counts, charges, licence and links to its records, download and receipt, …

fetch_order_recordsread-only

Returns a page of an order's delivered records (50 by default, up to 100) with the licence block on every page, as JSON or, on request, as CSV.

push_to_crm

Pushes an order's records to a connected CRM such as HubSpot or Salesforce, keyed on mydatahub_record_id, as record operations only: nothing is ever sent to th…

lookup_companyread-only

Looks up one company by Companies House number, domain or name and returns its Companies House fields and compliance facts, and for a signed-in caller the cont…

enrich_contact

Returns the full record for one known person, matched by email or by name with a company name, number or domain.

check_suppressionread-only

Checks the suppression status of up to 1,000 delivered records by record id, as of now, and names the lists that suppress each.

report_objection

Records an objection to direct marketing for one record, by record id or email hash.

get_compliance_receiptread-only

Returns an order's signed compliance receipt: an ES256 JWS with its decoded claims and a PDF link, covering records delivered, suppression removals, subscriber…

available_selection_categoriesread-only

Deprecated: lists the filter categories.

available_selection_valuesread-only

Deprecated: lists the values of one filter category by numeric id.

and 10 more

Test results

Safe mode: only tools labeled read-only were run. How the test works

!
Every tool is described
59 warnings
- !resolve_location: Inputs with no description: radius_miles
- !save_audience: Inputs with no description: name, brief
- !create_order: Inputs with no description: name
- !lookup_company: Inputs with no description: domain, name
- !enrich_contact: Inputs with no description: email, first_name, last_name, company_name, domain
- !check_suppression: Inputs with no description: record_ids
- !report_objection: Inputs with no description: channel, note
- !available_selection_values and search_discrete_values have nearly the same description (81% of their words are shared; the only difference: "list", "numeric", "take" vs "did", "mean", "near"), so an AI may not know which to pick
- and 32 more
note
!
Follows MCP best practices (not scored yet)
1 tip (not scored)
- !The tool list is about 294 KB (roughly 73k tokens). AI apps send it with every message, so it takes that much of the AI's memory before any work starts.
note
!
Resources and prompts work
1 warning (4 resources, 2 templates, 5 prompts)
- !refresh_saved_audience: Could not open it with sample inputs: audience_id: Expected an audience id: aud_ followed by a 26 character ULID in upper case Crockford base32, such as aud_01JB4H2K9V6XN3M9R6T1…
- and 5 more
note
✓
ConnectsConnected in 211 ms
pass
✓
Ready for the newest MCP versionUses the latest MCP version (2026-07-28)
pass
✓
Lists its toolsFound 34 tools
pass
✓
Tools are labeled safe or dangerousEvery tool has a safety label
pass
✓
No known hidden instructionsNone of the hidden-instruction wordings we know were found in tool descriptions. New wordings can get past this check.
pass
✓
Tools work with good inputs
All good
- ✓get_plans: Worked (491 ms)
- –get_account: Needs you to sign in, so it was not tested (not counted against the server)
- ✓list_filters: Worked (529 ms)
- ✓search_filter_values: Answered with an error, probably because of our sample input: dimension is not in the format this operation expects. Next step: Fix each entry in errors, using did_you_mean where given, and send it aga…
- ✓expand_job_titles: Answered with an error, probably because of our sample input: Send one of terms or bundle. Next step: Fix each entry in errors, using did_you_mean where given, and send it again. {"type":"https://email…
- ✓resolve_location: Worked (479 ms)
- ✓compile_audience: Worked (454 ms)
- ✓count_audience: Answered with an error, probably because of our sample input: Expected an audience id: aud_ followed by a 26 character ULID in upper case Crockford base32, such as aud_01JB4H2K9V6XN3M9R6T1Y5W8ZA. Next …
- and 26 more
pass
✓
Bad inputs give a clean error
All good
- ✓list_filters: Refused wrong type for "dataset" with a clear error
- ✓search_filter_values: Refused missing required input "dimension" with a clear error
- ✓search_filter_values: Refused wrong type for "dimension" with a clear error
- ✓expand_job_titles: Refused wrong type for "terms" with a clear error
- ✓resolve_location: Refused missing required input "text" with a clear error
- ✓resolve_location: Refused wrong type for "text" with a clear error
- ✓compile_audience: Refused missing required input "brief" with a clear error
- ✓compile_audience: Refused wrong type for "brief" with a clear error
- and 10 more
pass
✓
Fast enough
All good
- ✓Connecting took 211 ms
- ✓Listing tools took 678 ms
- ✓Average tool answer: 217 ms
- ✓list_filters: Slowest tool: list_filters (529 ms)
pass
✓
No secrets leakedChecked 67 error messages: no secrets or internal details
pass
–
Tools work togetherNo tool's answer could be fed into another tool, so there were no chains to test (not counted either way)
not tested

Test it yourself

Free and open source. Same checks, same score.

npx qamcp-cli https://mydatahub.emailmovers.com/mcp

Show this score on GitHub

Free for every server in the directory. Paste this into your README: the badge shows the score from our latest test.

[![QAmcp score](https://qamcp.dev/s/com-emailmovers-mydatahub.svg)](https://qamcp.dev/s/com-emailmovers-mydatahub.html)

Score over time

9 Nov 202689

One test so far. New tests will show here.

Reviews

Coming soon: reviews from people who use MyDataHub, signed in with GitHub, one per person. Reviews never change the score.

Is this your server? Get an alert the moment it breaks.

QAmcp Monitor tests it every hour and emails you when it goes down, the score drops, a tool changes or a secret leaks. Free for 1 server, no card.

Have a concern, or want it removed? Ask us and we'll take it off the list. How the directory works