commsharbor

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

37 tools

Its own safety labels

17 read-only, 2 labeled as deleting or destroying, 18 labeled as changing things

Memory it takes

About 7,300 tokens of your AI's memory, just for its tool descriptions

Largest answer

About 5.7 KB in our test

Tools that look alike

40 pairs of tools with nearly the same description, so the AI may pick the wrong one

Speed

Average answer in our test: 162 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 commsharbor MCP server safe?

• None of the hidden-instruction wordings we know were found in its 37 tool descriptions.
• Its error messages showed no secrets or internal details.
• 17 of its 37 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

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

What it can do

37 tools, in the server's own words.

and 13 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

  102 warnings

  • !commsharbor_email_domain_products_list and commsharbor_email_domain_products_add have nearly the same description (100% of their words are shared; the descriptions are identical), so an AI may not know which to pick
  • !commsharbor_email_mailbox_resume and commsharbor_email_alias_resume have nearly the same description (85% of their words are shared; the only difference: nothing vs "alia", "without", "changing"), so an AI may not know which to pick
  • !commsharbor_email_mailbox_resume and commsharbor_email_job_get have nearly the same description (84% of their words are shared; the only difference: "restore", "suspended" vs "job", "durable"), so an AI may not know which to pick
  • !commsharbor_email_mailbox_update and commsharbor_email_alias_update have nearly the same description (82% of their words are shared; the only difference: "name", "quota", "plan" vs "alia", "destination"), so an AI may not know which to pick
  • !commsharbor_email_alias_create and commsharbor_email_alias_update have nearly the same description (81% of their words are shared; the only difference: "create", "existing" vs "change", "destination", "within"), so an AI may not know which
  • !commsharbor_email_mailbox_password and commsharbor_email_mailbox_resume have nearly the same description (81% of their words are shared; the only difference: "set", "encrypted", "executor" vs "restore", "suspended"), so an AI may not know w
  • !commsharbor_email_mailbox_password and commsharbor_email_job_get have nearly the same description (81% of their words are shared; the only difference: "set", "encrypted", "executor" vs "job", "durable"), so an AI may not know which to pick
  • !commsharbor_email_mailbox_resume and commsharbor_email_alias_create have nearly the same description (81% of their words are shared; the only difference: "restore", "suspended" vs "create", "alia", "existing"), so an AI may not know which t
  • and 32 more
  note
• !
  Follows MCP best practices (not scored yet)

  2 tips (not scored)

  • !37 tools don't say whether they reach outside services like the web (openWorldHint): commsharbor_health, commsharbor_logout, commsharbor_me, commsharbor_context, commsharbor_organization_get, commsharbor_audit and 31 more
  • !4 tools answer with structured data but don't declare its shape (outputSchema): commsharbor_health, pricing, commsharbor_email_plans, commsharbor_brazil_email_report. If those answers change shape, nothing can catch it; with a schema, every
  note
• ✓
  ConnectsConnected in 800 ms
  pass
• ✓
  Lists its toolsFound 37 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

  • ✓commsharbor_health: Worked (168 ms)
  • –commsharbor_me: Needs you to sign in, so it was not tested (not counted against the server)
  • –commsharbor_context: Needs you to sign in, so it was not tested (not counted against the server)
  • –commsharbor_organization_get: Needs you to sign in, so it was not tested (not counted against the server)
  • –commsharbor_audit: Needs you to sign in, so it was not tested (not counted against the server)
  • –commsharbor_platform_context: Needs you to sign in, so it was not tested (not counted against the server)
  • ✓pricing: Worked (175 ms)
  • –commsharbor_email_google_catalog: Needs you to sign in, so it was not tested (not counted against the server)
  • and 29 more
  pass
• ✓
  Bad inputs give a clean error

  All good

  • ✓commsharbor_organization_get: Refused missing required input "organization_id" with a clear error
  • –commsharbor_organization_get: Needs you to sign in, so bad inputs were not tested
  • ✓commsharbor_audit: Refused missing required input "organization_id" with a clear error
  • –commsharbor_audit: Needs you to sign in, so bad inputs were not tested
  • ✓commsharbor_email_google_catalog: Refused missing required input "organization_id" with a clear error
  • –commsharbor_email_google_catalog: Needs you to sign in, so bad inputs were not tested
  • ✓commsharbor_email_domain_products_list: Refused missing required input "organization_id" with a clear error
  • –commsharbor_email_domain_products_list: Needs you to sign in, so bad inputs were not tested
  • and 12 more
  pass
• ✓
  Fast enough

  All good

  • ✓Connecting took 800 ms
  • ✓Listing tools took 150 ms
  • ✓Average tool answer: 162 ms
  • ✓pricing: Slowest tool: pricing (175 ms)
  pass
• ✓
  No secrets leakedChecked 63 error messages: no secrets or internal details
  pass
• ✓
  Resources and prompts workOpened 5 of 5 prompts
  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.

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.

Score over time

• 10 Nov 202689

One test so far. New tests will show here.

Reviews

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