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 / Opper

Opper

Q: Is Opper MCP server safe?

QAmcp's test on 11 Nov 2026: None of the hidden-instruction wordings we know were found in its 43 tool descriptions. Its error messages showed no secrets or internal details. 21 of its 43 tools are labeled read-only; the others may change or send things. Overall it scored 93/100 on 11 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 opper.ai · https://api.opper.ai/mcp

EU AI gateway: search 700+ models, create projects and API keys, manage rules, usage and traces.

43 tools✓ no account neededPricing: on their site

See pricing →Visit website

93 /100

PASS · WITH NOTES

12 of 12 checks · 11 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: 43 tools
Its own safety labels: 21 read-only, 16 labeled as deleting or destroying, 6 labeled as changing things
Memory it takes: About 31,600 tokens of your AI's memory, just for its tool descriptions
Largest answer: About 28.8 KB in our test
Speed: Average answer in our test: 165 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, 11 Nov 2026.

Is Opper MCP server safe?

None of the hidden-instruction wordings we know were found in its 43 tool descriptions.
Its error messages showed no secrets or internal details.
21 of its 43 tools are labeled read-only; the others may change or send things.
Overall it scored 93/100 on 11 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

43 tools, in the server's own words.

create_api_key

Mints a runtime API key for the project.

create_dynamic_route

Creates a route draft.

create_dynamic_route_deployment

Fail-closed deployment contract for infrastructure clients.

create_project

Creates a project in the calling organization.

create_rule

Creates a rule with a client-generated UUID.

delete_api_key

Permanently deletes a runtime API key.

delete_dynamic_route

Soft-deletes the route.

delete_project

Soft-deletes the project and permanently deletes every runtime API key bound to it.

delete_rule

Deletes a rule and stops its enforcement; inherited rules still apply.

deploy_dynamic_route

Validates the row-locked current draft and deploys it as a new immutable version, changing live routing.

get_dynamic_routeread-only

Returns the route, draft graph, validation issues and active version.

get_dynamic_route_deploymentread-only

Pure read of a completed deployment operation belonging to the same organization and authenticated actor.

get_dynamic_route_versionread-only

Returns a historical deployed version, including its graph and layout.

get_projectread-only

Fetches one project by UUID.

list_api_keysread-only

Lists the runtime API keys bound to a project.

list_dynamic_route_versionsread-only

Lists immutable deployed-version summaries for a route.

list_dynamic_routesread-only

Lists routes in the calling organization, including whether drafts differ from the deployed version.

list_projectsread-only

Lists every project in the calling organization.

list_rulesread-only

Lists rules in the calling organization.

restore_dynamic_route_version

Replaces the current draft with a historical version without changing live routing.

rollback_dynamic_route

Publishes a historical version as a new deployed version and changes live routing immediately.

simulate_dynamic_routeread-only

Omit version_number or set it to null to simulate the current draft, including an unpublished route.

undeploy_dynamic_route

Clears only the current active version pointer.

update_dynamic_route_draft

Saves draft changes without deploying.

and 19 more

Test results

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

!
Every tool is described
5 warnings
- !create_api_key: Inputs with no description: kind, name
- !create_project: Inputs with no description: description, name
- !create_rule: Inputs with no description: kind
- !update_project: Inputs with no description: description, name
- !create_organization_personal_key: Inputs with no description: kind, name
note
!
Follows MCP best practices (not scored yet)
1 tip (not scored)
- !The tool list is about 188 KB (roughly 47k tokens). AI apps send it with every message, so it takes that much of the AI's memory before any work starts.
note
✓
ConnectsConnected in 760 ms
pass
✓
Ready for the newest MCP versionUses the latest MCP version (2026-07-28)
pass
✓
Login setup follows the spec
All good
- ✓The server asks for a login (it answers 401 to an anonymous request).
- ✓The 401 reply points AI apps to the login details (resource_metadata).
- ✓The login details name this exact server.
- ✓Found the login server's description at https://api.opper.ai/.well-known/oauth-authorization-server/mcp/oauth.
- ✓The login server's name matches what the server points to.
- ✓PKCE with S256 is supported.
- ✓AI apps can register themselves (dynamic registration and client ID metadata documents).
pass
✓
Lists its toolsFound 43 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_dynamic_route: Needs you to sign in, so it was not tested (not counted against the server)
- –get_dynamic_route_deployment: Needs you to sign in, so it was not tested (not counted against the server)
- –get_dynamic_route_version: Needs you to sign in, so it was not tested (not counted against the server)
- ✓get_project: Answered with an error, probably because of our sample input: argument "uuid" must be a UUID
- ✓list_api_keys: Answered with an error, probably because of our sample input: argument "uuid" must be a UUID
- –list_dynamic_route_versions: Needs you to sign in, so it was not tested (not counted against the server)
- –list_dynamic_routes: Needs you to sign in, so it was not tested (not counted against the server)
- –list_projects: Needs you to sign in, so it was not tested (not counted against the server)
- and 32 more
pass
✓
Bad inputs give a clean error
All good
- ✓get_dynamic_route: Refused missing required input "name" with a clear error
- –get_dynamic_route: Needs you to sign in, so bad inputs were not tested
- ✓get_dynamic_route_deployment: Refused missing required input "idempotency_key" with a clear error
- –get_dynamic_route_deployment: Needs you to sign in, so bad inputs were not tested
- ✓get_dynamic_route_version: Refused missing required input "name" with a clear error
- –get_dynamic_route_version: Needs you to sign in, so bad inputs were not tested
- ✓get_project: Refused missing required input "uuid" with a clear error
- ✓get_project: Refused wrong type for "uuid" with a clear error
- and 13 more
pass
✓
Fast enough
All good
- ✓Connecting took 760 ms
- ✓Listing tools took 806 ms
- ✓Average tool answer: 165 ms
- ✓list_models: Slowest tool: list_models (172 ms)
pass
✓
No secrets leakedChecked 63 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://api.opper.ai/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/ai-opper-opper.svg)](https://qamcp.dev/s/ai-opper-opper.html)

Score over time

11 Nov 202693

One test so far. New tests will show here.

Reviews

Coming soon: reviews from people who use Opper, 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