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 / Agent Coach (Upskill)

Agent Coach (Upskill)

Q: Is Agent Coach (Upskill) MCP server safe?

QAmcp's test on 10 Nov 2026: None of the hidden-instruction wordings we know were found in its 13 tool descriptions. Its error messages showed no secrets or internal details. 6 of its 13 tools are labeled read-only; the others may change or send things. Overall it scored 89/100 on 10 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 upskill.coach · https://upskill.coach/mcp

Free coach for AI agents: security-vetted skills, MCP servers and prompts, plus daily check-ins.

13 tools✓ no account neededPricing: on their site

See pricing →Visit website

89 /100

PASS · WITH NOTES

12 of 12 checks · 10 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: 13 tools
Its own safety labels: 6 read-only, 1 labeled as deleting or destroying, 6 labeled as changing things
Memory it takes: About 6,000 tokens of your AI's memory, just for its tool descriptions
Largest answer: About 17.8 KB in our test
Speed: Average answer in our test: 615 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, 10 Nov 2026.

Is Agent Coach (Upskill) MCP server safe?

None of the hidden-instruction wordings we know were found in its 13 tool descriptions.
Its error messages showed no secrets or internal details.
6 of its 13 tools are labeled read-only; the others may change or send things.
Overall it scored 89/100 on 10 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

13 tools, in the server's own words.

register_agent

Use this first, once.

set_goals

Use this after registering (and whenever goals change).

get_recommendationsread-only

Use this after set_goals.

update_items

Use this after your owner approves an item (action=adopt), rejects one (decline), when an item doesn't fit your job (not_relevant: not a judgment on its qualit…

check_in

Use this once a day.

get_reportread-only

Use this to see your KPIs vs baseline, every item's verdict with evidence, and check-in history.

share_report

Use this only with your owner's OK.

get_digestread-only

Use this to preview the daily owner digest (subject, html, text).

rotate_key

Use this if your key may have leaked.

delete_agent

Use this only when your owner asks.

get_rulesread-only

Use this to see exactly how the coach decides item verdicts (permanent / promising / temporary / trial / unproven / pending_removal), signals, and overall heal…

search_libraryread-only

Use this to find skills, MCP servers, prompts or repos by keyword outside your matched recommendations.

get_library_itemread-only

Use this to fetch one vetted library item by slug: install, source, license and (when the license allows) the full SKILL.md text.

Test results

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

!
Ready for the newest MCP version
1 warning
- !Uses an older MCP version (2025-11-25). It works today, but it isn't on the July 2026 version (2026-07-28) yet.
note
!
Every tool is described
7 warnings
- !set_goals: Inputs with no description: job_type, stage, job_title, tools, platform, tags
- !get_recommendations: Inputs with no description: limit
- !update_items: Inputs with no description: item_ids, reason
- !check_in: Inputs with no description: releases, owner_blockers, owner_corrections, interruptions, incidents, items, tasks, owner_feedback, notes
- !share_report: Inputs with no description: public
- !search_library: Inputs with no description: kind, limit
- !get_library_item: Inputs with no description: slug
note
✓
ConnectsConnected in 753 ms
pass
✓
Lists its toolsFound 13 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_recommendations: Needs you to sign in, so it was not tested (not counted against the server)
- –get_report: Needs you to sign in, so it was not tested (not counted against the server)
- –get_digest: Needs you to sign in, so it was not tested (not counted against the server)
- ✓get_rules: Worked (61 ms)
- ✓search_library: Worked (591 ms)
- ✓get_library_item: Answered with an error, probably because of our sample input: {"error":{"code":"not_found","message":"Unknown or unvetted item"}} {"error":{"code":"not_found","message":"Unknown or unvetted item"}}
- –register_agent: Not run: labeled as changing data
- –set_goals: Not run: labeled as changing data
- and 5 more
pass
✓
Bad inputs give a clean error
All good
- –get_recommendations: Needs you to sign in, so bad inputs were not tested
- –get_report: Needs you to sign in, so bad inputs were not tested
- –get_digest: Needs you to sign in, so bad inputs were not tested
- ✓search_library: Refused wrong type for "query" with a clear error
- ✓get_library_item: Refused missing required input "slug" with a clear error
- ✓get_library_item: Refused wrong type for "slug" with a clear error
pass
✓
Tools work togethersearch_library → get_library_item: get_library_item found what search_library returned (slug = "obra-superpowers-brainstorming")
pass
✓
Fast enough
All good
- ✓Connecting took 753 ms
- ✓Listing tools took 67 ms
- ✓Average tool answer: 615 ms
- ✓get_library_item: Slowest tool: get_library_item (1.2 s)
pass
✓
No secrets leakedChecked 11 error messages: no secrets or internal details
pass
✓
Follows MCP best practices (not scored yet)Tool names, safety labels and tool list size follow the MCP guide
pass

Test it yourself

Free and open source. Same checks, same score.

npx qamcp-cli https://upskill.coach/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/coach-upskill-agent-coach.svg)](https://qamcp.dev/s/coach-upskill-agent-coach.html)

Score over time

10 Nov 202689

One test so far. New tests will show here.

Reviews

Coming soon: reviews from people who use Agent Coach (Upskill), 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