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 / CallMCP — Phone Call MCP Server

CallMCP — Phone Call MCP Server

Q: Is CallMCP — Phone Call MCP Server server safe?

QAmcp's test on 12 Nov 2026: None of the hidden-instruction wordings we know were found in its 194 tool descriptions. Its error messages showed no secrets or internal details. 13 of its 60 tools are labeled read-only; the others may change or send things. Overall it scored 86/100 on 12 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 callmcp.ai · https://callmcp.ai/mcp

Phone call MCP server: AI agents place and answer real calls, text, email and WhatsApp via KaiCalls.

194 tools✓ no account neededPricing: on their site

See pricing →Visit website

86 /100

PASS · WITH NOTES

Partial: 10 of 12 checks · 12 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: 194 tools
Its own safety labels: 66 read-only, 75 labeled as deleting or destroying, 53 labeled as changing things
Memory it takes: About 60,900 tokens of your AI's memory, just for its tool descriptions

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, 12 Nov 2026.

Is CallMCP — Phone Call MCP Server server safe?

None of the hidden-instruction wordings we know were found in its 194 tool descriptions.
Its error messages showed no secrets or internal details.
13 of its 60 tools are labeled read-only; the others may change or send things.
Overall it scored 86/100 on 12 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

None found in GitHub Security Advisories and OSV.dev on 12 Nov 2026.

That means none were published there, not that there are no problems. QAmcp tests how a server behaves. It does not scan source code.

What it can do

60 tools, in the server's own words.

request_human_approval

Ask a named principal for a specific decision using the provisioned voice approval channel.

connect_agent

Connect an external agent to a business and its signed event outbox.

get_agent_connectionread-only

Discover the external agent connection, signed webhook configuration, scoped API operation catalog, channel readiness and remaining runtime requirements.

disconnect_agent

Disconnect the external agent and disable its webhook atomically.

get_lsa_account_reportread-only

Read the authorized advertiser's live Google Local Services aggregate account report, including actual spend, currency and charged lead totals.

get_lsa_setupread-only

Inspect Google LSA account selection and sync freshness, with setup steps.

sync_lsa

Sync an explicitly selected connected LSA advertiser.

list_lsa_leadsread-only

List synced Google LSA leads, charge and credit states, source IDs and rating details.

get_lsa_conversationsread-only

Read the latest 200 synced Google LSA conversation events for an LSA lead, including customer and advertiser messages.

get_lsa_reportread-only

Report LSA lead counts, types, observed statuses, ratings and credit counts for an explicit UTC window (start inclusive, end exclusive).

send_lsa_reply

Send an authorized text reply inside an existing Google LSA message lead through the Ads API.

rate_lsa_lead

Submit an authorized, truthful Google LSA quality survey and return Google's actual bonus-credit decision.

get_callrailread-only

Read a business's CallRail call-tracking setup.

manage_callrail

Connect CallRail call tracking to a business's line.

agency_profile_bootstrap

Registers a new agency for the signed-in user and makes them its owner, using the given brand and support details.

agency_profile_update

Replaces the agency brand, support details and requested commercial mode.

agency_profile_invite

Creates a staff invitation for an email address with the admin, operator or reader role and returns a one-time invite link that expires in 7 days.

agency_profile_accept_invite

Joins the agency using the token from an invite link.

agency_profile_revoke_invite

Cancels a pending staff invitation so its link can no longer be used.

agency_profile_assign

Gives an active agency staff member access to one client business with the admin or member business role.

agency_profile_unassign

Removes a staff member's agency assignment to one client business, and removes the business membership if this assignment created it.

agency_profile_revoke_member

Removes a staff member from the agency: revokes their membership, all their client assignments, their accepted invites and every API key issued to them through…

agency_profile_issue_key

Creates a new API key for yourself in this agency, limited to the requested scopes, and returns the secret once.

agency_offer_save

Creates a new offer draft, or replaces an existing draft when offerId and its current revision are given.

and 36 more

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 (2025-06-18). It works today, but it isn't on the July 2026 version (2026-07-28) yet.
note
!
Every tool is described
57 warnings
- !request_human_approval: Inputs with no description: source_request_id, source_intent_id, message_id, responds_to_hash, principal_id, summary, risk_level, required_factors, expires_at, callback_secret_id
- !connect_agent: Inputs with no description: business_id, agent_id, external_agent_id, webhook_url, expected_revision, external_reply_timeout_seconds
- !get_agent_connection: Inputs with no description: business_id
- !disconnect_agent: Inputs with no description: business_id, expected_revision
- !get_lsa_account_report: Inputs with no description: business_id, connection_id, start_date, end_date
- !get_lsa_setup: Inputs with no description: business_id
- !sync_lsa: Inputs with no description: business_id, connection_id
- !list_lsa_leads: Inputs with no description: business_id, offset, limit
- and 32 more
note
!
Follows MCP best practices (not scored yet)
1 tip (not scored)
- !The tool list is about 363 KB (roughly 91k 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 598 ms
pass
✓
Lists its toolsFound 194 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
✓
Fast enough
All good
- ✓Connecting took 598 ms
- ✓Listing tools took 201 ms
pass
✓
No secrets leakedChecked 147 error messages: no secrets or internal details
pass
✓
Resources and prompts work
All good (1 resource)
- ✓Opened 1 of 1 resources
- ✓An address that does not exist gets a clean "not found" error
pass
–
Tools work with good inputs
Not tested
- –get_agent_connection: Needs you to sign in, so it was not tested (not counted against the server)
- –get_lsa_account_report: Needs you to sign in, so it was not tested (not counted against the server)
- –get_lsa_setup: Needs you to sign in, so it was not tested (not counted against the server)
- –list_lsa_leads: Needs you to sign in, so it was not tested (not counted against the server)
- –get_lsa_conversations: Needs you to sign in, so it was not tested (not counted against the server)
- –get_lsa_report: Needs you to sign in, so it was not tested (not counted against the server)
- –get_callrail: Needs you to sign in, so it was not tested (not counted against the server)
- –agency_offer_preview: Needs you to sign in, so it was not tested (not counted against the server)
- and 32 more
not tested
–
Bad inputs give a clean error
Not tested
- –get_agent_connection: Needs you to sign in, so bad inputs were not tested
- –get_lsa_account_report: Needs you to sign in, so bad inputs were not tested
- –get_lsa_setup: Needs you to sign in, so bad inputs were not tested
- –list_lsa_leads: Needs you to sign in, so bad inputs were not tested
- –get_lsa_conversations: Needs you to sign in, so bad inputs were not tested
- –get_lsa_report: Needs you to sign in, so bad inputs were not tested
- –get_callrail: Needs you to sign in, so bad inputs were not tested
- –agency_offer_preview: Needs you to sign in, so bad inputs were not tested
- and 32 more
not tested
–
Tools work togetherNo safe tool returned data to start a chain from, so chains were not tested (not counted either way)
not tested

Test it yourself

Free and open source. Same checks, same score.

npx qamcp-cli https://callmcp.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-callmcp-server.svg)](https://qamcp.dev/s/ai-callmcp-server.html)

Score over time

12 Nov 202686

One test so far. New tests will show here.

Reviews

Coming soon: reviews from people who use CallMCP — Phone Call MCP Server, 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