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 / goldseam

goldseam

Q: Is goldseam MCP server safe?

QAmcp's test on 8 Nov 2026: None of the hidden-instruction wordings we know were found in its 70 tool descriptions. Its error messages showed no secrets or internal details. 56 of its 60 tools are labeled read-only; the others may change or send things. Overall it scored 70/100 on 8 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 ksaworks.com · https://goldseam.ksaworks.com/mcp

The U.S. Marketplace of Learning2Earning Opportunity | Sourced checklists, progress, prior learning

70 tools✓ no account neededPricing: not listed

Visit website

70 /100

FAIL

13 of 13 checks · 8 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: 70 tools
Its own safety labels: 65 read-only, 5 labeled as changing things
Memory it takes: About 32,100 tokens of your AI's memory, just for its tool descriptions
Largest answer: About 334.4 KB in our test
Tools that look alike: 30 pairs of tools with nearly the same description, so the AI may pick the wrong one
Speed: Average answer in our test: 519 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, 8 Nov 2026.

Is goldseam MCP server safe?

None of the hidden-instruction wordings we know were found in its 70 tool descriptions.
Its error messages showed no secrets or internal details.
56 of its 60 tools are labeled read-only; the others may change or send things.
Overall it scored 70/100 on 8 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

Not looked up yet. QAmcp tests how a server behaves. It does not scan source code.

What it can do

60 tools, in the server's own words.

search_certificationsread-only

Which certification is this prose name?

get_certificationread-only

Everything held about one certification: issuer and issuer domain, occupations served, exam and renewal requirements, education-or-experience routes, Credentia…

certifications_for_occupationread-only

What certifications serve this occupation?

occupations_for_certificationread-only

What occupations does this certification serve?

related_certificationsread-only

What else serves the same occupation?

certifications_for_programread-only

Broad academic cross-reference by instructional program (CIP).

list_issuersread-only

The issuing organizations, optionally filtered by a name query.

resolve_termread-only

Where does a word place in the KSA model?

expand_termread-only

What is this word, upward?

get_dictionary_summaryread-only

What does the KSA Dictionary hold?

crosswalk_coderead-only

What does this code cross to?

open_coach

Open a visible, speaking coach (coach, advisor, counselor, teacher, mentor or another role) in this chat as an MCP App view, for any GoldSeam service.

coach_say

Speak through the open coach: the words, queued for the coach and spoken in the voice the person chose on the coach once the coach view holds the coach; and op…

coach_listen

Listen for the person's next spoken reply to the open coach, through the coach session open_coach started: start in the turn that opened the coach, with sessio…

coach_board

Show the whiteboard beside the open coach, or close it (show false).

compare_program_checklistsread-only

Compare what a program takes at several institutions: for each IPEDS UNITID, that institution's published checklist for the program (requirement groups, items,…

find_institutionread-only

Which institution is this?

programs_for_institutionread-only

What programs does this institution publish?

get_checklistread-only

What does this award require?

learning_unitsread-only

What courses does this institution publish?

get_institutionread-only

Everything held about one institution: its identity row, identifiers, addresses, websites, accreditations, relationships, identity history, published costs and…

find_catalogsread-only

Which published catalog editions are held for this institution?

get_catalogread-only

One catalog edition: its row, the institutions that use it (and whom it is shared from), and how many programs it holds.

find_programsread-only

Which programs match these words, across institutions?

and 36 more

Test results

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

✕
Tools work with good inputs
5 failed, 6 warnings
- !compare_program_checklists: Answered with about 161 KB of text. Many AI apps cut off tool answers this big, so the AI may never see the end of it.
- !programs_for_institution: Answered with about 135 KB of text. Many AI apps cut off tool answers this big, so the AI may never see the end of it.
- !get_learning_unit: Answered with about 334 KB of text. Many AI apps cut off tool answers this big, so the AI may never see the end of it.
- !find_checklists: Answered with about 121 KB of text. Many AI apps cut off tool answers this big, so the AI may never see the end of it.
- ✕checklists_requiring_learning_unit: No answer within the time limit
- !search_programs_and_learning_units: Answered with about 136 KB of text. Many AI apps cut off tool answers this big, so the AI may never see the end of it.
- and 34 more
fail
!
Every tool is described
146 warnings
- !coach_say: Inputs with no description: cues
- !my_selfie_ksa_view: Inputs with no description: language
- !draft_my_claims: Inputs with no description: language
- !suggest_credit_for_my_learning: Inputs with no description: language
- !add_to_my_selfie: Inputs with no description: descriptions, skip_topics
- !respond_to_my_statements: Inputs with no description: responses
- !my_selfie_checkpoint: Inputs with no description: note
- !read_my_document_for_selfie: Inputs with no description: files, pasted
- and 32 more
note
!
Bad inputs give a clean error
16 warnings
- !find_institution: Accepted wrong type for "city" without complaining
- !programs_for_institution: Accepted wrong type for "unitid" without complaining
- !find_catalogs: Accepted wrong type for "unitid" without complaining
- !find_policies: Accepted wrong type for "unitid" without complaining
- and 36 more
note
!
Fast enough
1 warning
- !search_programs_and_learning_units: Slowest tool: search_programs_and_learning_units (8.9 s): users will notice the wait
- and 3 more
note
!
Follows MCP best practices (not scored yet)
3 tips (not scored)
- !5 tools not labeled read-only don't say whether running them twice is safe (idempotentHint): open_coach, coach_say, coach_listen, coach_board, mail_my_bio
- !65 tools don't say whether they reach outside services like the web (openWorldHint): search_certifications, get_certification, certifications_for_occupation, occupations_for_certification, related_certifications, certifications_for_program
- !The tool list is about 171 KB (roughly 43k 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 (6 resources, 8 prompts)
- !Listing resource templates failed: Error POSTing to endpoint: {"error":{"code":-32601,"message":"Method not found: resources/templates/list"},"id":173,"jsonrpc":"2.0"}
- and 3 more
note
✓
ConnectsConnected in 212 ms
pass
✓
Ready for the newest MCP versionUses the latest MCP version (2026-07-28)
pass
✓
Lists its toolsFound 70 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 together
5 chains working
- ✓search_certifications → get_certification: get_certification found what search_certifications returned (id = "LU_CERT_ONET_29117100_99C6BEF356")
- ✓search_certifications → occupations_for_certification: occupations_for_certification found what search_certifications returned (id = "LU_CERT_ONET_29117100_99C6BEF356")
- ✓search_certifications → related_certifications: related_certifications found what search_certifications returned (id = "LU_CERT_ONET_29117100_99C6BEF356")
- ✓get_certification → occupations_for_certification: occupations_for_certification found what get_certification returned (id = "LU_CERT_ONET_13108200_8F3480AC7A")
- ✓get_certification → related_certifications: related_certifications found what get_certification returned (id = "LU_CERT_ONET_13108200_8F3480AC7A")
pass
✓
No secrets leakedChecked 102 error messages: no secrets or internal details
pass

Test it yourself

Free and open source. Same checks, same score.

npx qamcp-cli https://goldseam.ksaworks.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-ksaworks-goldseam.svg)](https://qamcp.dev/s/com-ksaworks-goldseam.html)

Score over time

8 Nov 202670

One test so far. New tests will show here.

Reviews

Coming soon: reviews from people who use goldseam, 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