# BumbleBox (remote · api.bumblebox.ai)

Search and triage all your mailboxes; draft replies and campaigns; send only as you allow.

- Trust score: 81/100 (high trust)
- Change this week: +3
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-04

## Components

- remote · `api.bumblebox.ai`: 81/100 (this document), [markdown](https://verifymcp.io/servers/ai-bumblebox-mail/api-mcp.md), [page](https://verifymcp.io/servers/ai-bumblebox-mail/api-mcp)

## Channel facts

- Endpoint: `https://api.bumblebox.ai/api/mcp`
- Transports: `streamable-http`
- Auth: `required`
- Version: `2.1.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-04.

- **Endpoint Security**: 89/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation is enforced on tool calls, advertised via RFC 9728 protected-resource metadata. Discovery is public, which costs nothing: no tool can be invoked without a token.
  - HTTPS is enforced; there's no plaintext access path.
  - HSTS check failed: the Strict-Transport-Security header is absent.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
  - The authorisation server offers only Dynamic Client Registration (RFC 7591), which MCP 2026-07-28 deprecated in favour of Client ID Metadata Documents.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 70/100
  - AI-judged instruction clarity (good).
  - Tool/resource definitions use about 1424 tokens (~83/item across 17 items; 17 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 43/100
  - Stability observed for 13 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 86/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 50% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - All 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.
  - An AI judge read all 18 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the BumbleBox MCP server?

BumbleBox is a hosted endpoint at https://api.bumblebox.ai/api/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http ai-bumblebox-mail 'https://api.bumblebox.ai/api/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "ai-bumblebox-mail": {
      "url": "https://api.bumblebox.ai/api/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "ai-bumblebox-mail": {
      "type": "http",
      "url": "https://api.bumblebox.ai/api/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.ai-bumblebox-mail]
url = "https://api.bumblebox.ai/api/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "ai-bumblebox-mail": {
      "type": "remote",
      "url": "https://api.bumblebox.ai/api/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add ai-bumblebox-mail --url 'https://api.bumblebox.ai/api/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  ai-bumblebox-mail:
    url: "https://api.bumblebox.ai/api/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "ai-bumblebox-mail": {
      "Transport": "http",
      "Url": "https://api.bumblebox.ai/api/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add ai-bumblebox-mail -t streamable-http -u 'https://api.bumblebox.ai/api/mcp'
```

### Other

```json
{
  "mcpServers": {
    "ai-bumblebox-mail": {
      "type": "http",
      "url": "https://api.bumblebox.ai/api/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-03 (score 81, +1)

No change was recorded against any check on this day. Stability & Change Management went from 37 to 40. That category is still filling its 30-day observation window: 11 days of observed history at the previous scan, 12 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-01 (score 80, +1)

No change was recorded against any check on this day. Stability & Change Management went from 30 to 33. That category is still filling its 30-day observation window: 9 days of observed history at the previous scan, 10 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-30 (score 79, +1)

No change was recorded against any check on this day. Stability & Change Management went from 23 to 27. That category is still filling its 30-day observation window: 7 days of observed history at the previous scan, 8 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-28 (score 78, +1)

No change was recorded against any check on this day. Stability & Change Management went from 17 to 20. That category is still filling its 30-day observation window: 5 days of observed history at the previous scan, 6 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-26 (score 77, +1)

- [security] Tool “draft_reply” rewrote its description, which is the text the model reads
- [security] Tool “search_emails” rewrote its description, which is the text the model reads
- [security] Tool “send_reply” rewrote its description, which is the text the model reads

### 2026-10-23 (score 76, +1)

- [functional improvement] Stability: unverified → 0.03

### 2026-10-22 (score 75)

First indexed and scored.

## MCP tools (17)

### `search_emails` (~83 tokens)

Search emails

Searches every connected mailbox (sender, subject, preview), newest first. Operators: from:, subject:, before:YYYY-MM-DD, after:YYYY-MM-DD, in:sent (only what the user sent, from any mailbox or alias); the rest is free text. Deleted emails are not returned.

Input parameters:

- `limit` (integer)
- `query` (string, required)

Output parameters:

- `results` (array)

### `get_email` (~54 tokens)

Read an email

One email in full: sender, recipients, date, subject, plain-text body, attachment names, folder and BumbleBox classification.

Input parameters:

- `id` (string, required): Email id, as returned by search_emails or list_to_answer.

Output parameters:

- `email` (object)

### `list_to_answer` (~34 tokens)

Emails waiting for a reply

Received emails BumbleBox classified as needing a reply and still in the inbox, newest first.

Input parameters:

- `limit` (integer)

Output parameters:

- `emails` (array)

### `search_contacts` (~57 tokens)

Search contacts

Contacts BumbleBox knows from the mail history (name, address, organization domain, exchange counts, last exchanges, VIP mark, note), matched on name, address or domain.

Input parameters:

- `limit` (integer)
- `query` (string)

Output parameters:

- `contacts` (array)

### `list_folders` (~31 tokens)

List mail folders

The folders of each connected mailbox, as last synced. Trash, spam, sent and drafts are left out.

Output parameters:

- `mailboxes` (array)

### `list_calendar_events` (~47 tokens)

List calendar events

Events of the connected Google calendars between two dates (default: the next 7 days, at most 62 days).

Input parameters:

- `from` (string)
- `to` (string)

Output parameters:

- `events` (array)

### `draft_reply` (~137 tokens)

Prepare a reply (not sent)

Saves the text as the reply draft of a received email, from the address the email was written to (a brand address or the mailbox). needsIdentity: that address cannot send until the user verifies it in BumbleBox. Nothing is sent: the draft waits in BumbleBox (Pilotage › prepared replies) until the user reviews it and clicks Send. Saving again replaces the previous draft of the same email.

Input parameters:

- `body` (string, required): Plain-text reply body.
- `email_id` (string, required): Email id, as returned by search_emails or list_to_answer.
- `subject` (string): Defaults to "Re: <original subject>".

Output parameters:

- `draft` (object)

### `send_reply` (~168 tokens)

Reply to the sender

Replies to the sender of one received email, and only to them (no Cc, no Bcc, no attachment), from the address the email was written to. What happens next is the level the user set in BumbleBox for this kind of reply: a draft, a reply waiting for one tap (the default), or a reply that leaves after a delay the user can cancel. Anything that commits the user (amounts, bank details, acceptance, contracts, dates, links) always waits for the user, and so does a reply from an address the user has not verified (needsIdentity). The result says which one applied.

Input parameters:

- `body` (string, required): Plain-text reply body.
- `email_id` (string, required): Email id, as returned by search_emails or list_to_answer.

Output parameters:

- `reply` (object)

### `archive_email` (~58 tokens)

Archive an email

Moves one inbox email to its mailbox's archive folder, on the mail server too. Undone by restore_to_inbox. Nothing is deleted.

Input parameters:

- `email_id` (string, required): Email id, as returned by search_emails or list_to_answer.

Output parameters:

- `moved` (object)

### `move_email` (~73 tokens)

Move an email to a folder

Moves one inbox email to an existing folder of the same mailbox (see list_folders), on the mail server too. No folder is created. Undone by restore_to_inbox.

Input parameters:

- `email_id` (string, required): Email id, as returned by search_emails or list_to_answer.
- `folder` (string, required)

Output parameters:

- `moved` (object)

### `trash_email` (~115 tokens)

Put an email in the trash

Moves one inbox email to its mailbox's trash when BumbleBox certifies it as noise (spam, newsletter, promotion, with a unanimous safety vote and no known contact or action word). Any other email is not moved: it becomes a proposal the user confirms in one tap in BumbleBox. Never a permanent deletion; undone by restore_to_inbox.

Input parameters:

- `email_id` (string, required): Email id, as returned by search_emails or list_to_answer.
- `reason` (string): Why, shown to the user with a proposal.

Output parameters:

- `result` (object)

### `restore_to_inbox` (~51 tokens)

Put an email back in the inbox

Moves an archived, filed or trashed email back to the inbox, on the mail server too.

Input parameters:

- `email_id` (string, required): Email id, as returned by search_emails or list_to_answer.

Output parameters:

- `email_id` (string)
- `restored` (number)

### `list_missions` (~43 tokens)

List campaigns

The user's prospection campaigns (missions), most recently active first, with the number of drafts waiting, emails sent and prospects who replied.

Input parameters:

- `limit` (integer)

Output parameters:

- `missions` (array)

### `list_campaign_drafts` (~69 tokens)

List the drafts of a campaign

The drafts of one campaign that wait for the user to validate them in BumbleBox (recipient, subject, plain-text body, sender), newest first, with the true totals.

Input parameters:

- `limit` (integer)
- `mission_id` (string, required): Campaign id, as returned by list_missions.

Output parameters:

- `counts` (object)
- `drafts` (array)

### `list_replies_to_validate` (~52 tokens)

Campaign replies waiting for the user

Replies received on campaigns that still need the user: prepared reply drafts waiting for validation, and replies with no answer yet (with the stored AI triage: kind, summary, whether a human is needed).

Output parameters:

- `drafts` (array)
- `pending`

### `create_campaign_draft` (~110 tokens)

Write a campaign draft (not sent)

Writes one draft inside an existing campaign, to one recipient (added to the campaign list if new), from the campaign's sender. Nothing is sent: the draft waits for the user to validate the campaign send in BumbleBox.

Input parameters:

- `body` (string, required): Plain-text body.
- `mission_id` (string, required): Campaign id, as returned by list_missions.
- `name` (string): Recipient name, for a new prospect.
- `subject` (string, required)
- `to` (string, required)

Output parameters:

- `draft` (object)

### `update_campaign_draft` (~87 tokens)

Edit a campaign draft (not sent)

Changes the subject, body or recipient of a draft that is still a draft. Nothing is sent, and a draft the user already validated cannot be changed.

Input parameters:

- `body` (string)
- `draft_id` (string, required)
- `mission_id` (string, required): Campaign id, as returned by list_missions.
- `subject` (string)
- `to` (string)

Output parameters:

- `draft` (object)

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/ai-bumblebox-mail/api-mcp#diagnostics

## Score history

- 2026-11-04: 81
- 2026-11-03: 81
- 2026-11-02: 80
- 2026-11-01: 80
- 2026-10-31: 79
- 2026-10-30: 79
- 2026-10-29: 78
- 2026-10-28: 78
- 2026-10-27: 77
- 2026-10-26: 77
- 2026-10-25: 76
- 2026-10-24: 76
- 2026-10-23: 76
- 2026-10-22: 75

## Common questions

### What is the BumbleBox MCP server?

BumbleBox is an MCP server listed in the public MCP registry as ai.bumblebox/mail. Search and triage all your mailboxes; draft replies and campaigns; send only as you allow. This page covers its hosted endpoint (https://api.bumblebox.ai/api/mcp).

### Is the BumbleBox MCP server safe to use?

BumbleBox scores 81 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the BumbleBox MCP server expose?

BumbleBox exposes 17 tools: search_emails, get_email, list_to_answer, search_contacts, list_folders, and 12 more. Their descriptions and schemas cost roughly 1,269 tokens of context every time the server is loaded.

### Does the BumbleBox MCP server require authentication?

Yes. BumbleBox asked us for credentials when we connected, so you will need to authorise it in your MCP client before it can do anything.

### Is the BumbleBox MCP server still maintained?

BumbleBox is still listed as active in the MCP registry. We last reached this channel on 4 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

## Links

- Remote endpoint: https://api.bumblebox.ai/api/mcp
- Repository: https://github.com/BumbleBox/bumblebox-mcp
- Website: https://bumblebox.ai/integrations
- Changelog RSS feed: https://verifymcp.io/servers/ai-bumblebox-mail/api-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/ai-bumblebox-mail/api-mcp.json
- HTML version of this page: https://verifymcp.io/servers/ai-bumblebox-mail/api-mcp