# SMB Sandbox (npm · @teamshift/sandbox-mcp)

Mock small-business CRM, inbox, calendar, invoicing and phone tools for testing AI agents safely.

- Trust score: 79/100 (medium)
- Change this week: +3
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-10-23

## Components

- npm · `@teamshift/sandbox-mcp`: 79/100 (this document), [markdown](https://verifymcp.io/servers/teamshift-io-sandbox-mcp/teamshift-sandbox-mcp.md), [page](https://verifymcp.io/servers/teamshift-io-sandbox-mcp/teamshift-sandbox-mcp)

## Channel facts

- Registry: `npm`
- Package: `@teamshift/sandbox-mcp`
- Version: `0.1.0`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, 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 2026-10-23.

- **Supply Chain Security**: 98/100
  - No malware found by supply-chain analysis.
  - No known CVEs affecting this package version or its production dependencies.
  - No install/post-install scripts declared.
  - 30 of 93 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 48/100
  - Source repository is publicly reachable at the declared URL.
  - Provenance check failed: no build-provenance attestation is published.
  - Clear OSI-approved license (Apache-2.0).
  - Actively maintained (last published 19 days ago).
  - Publishes a security disclosure policy (SECURITY.md).
- **Schema Quality & AI Usability**: 84/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (good).
  - Tool/resource definitions use about 3451 tokens (~71/item across 48 items; 46 tools + 2 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 63/100
  - Stability observed for 19 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 80/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 40% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - All 5 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.
  - An AI judge read all 48 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 SMB Sandbox MCP server?

SMB Sandbox runs locally as an npm package, launched with npx -y @teamshift/sandbox-mcp. 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 teamshift-io-sandbox-mcp -- npx -y @teamshift/sandbox-mcp
```

### Cursor

```json
{
  "mcpServers": {
    "teamshift-io-sandbox-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@teamshift/sandbox-mcp"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "teamshift-io-sandbox-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@teamshift/sandbox-mcp"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add teamshift-io-sandbox-mcp -- npx -y @teamshift/sandbox-mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "teamshift-io-sandbox-mcp": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "@teamshift/sandbox-mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add teamshift-io-sandbox-mcp --command npx --arg -y --arg @teamshift/sandbox-mcp
```

### Hermes

```yaml
mcp_servers:
  teamshift-io-sandbox-mcp:
    command: "npx"
    args: ["-y", "@teamshift/sandbox-mcp"]
```

### Netclaw

```json
{
  "McpServers": {
    "teamshift-io-sandbox-mcp": {
      "Transport": "stdio",
      "Command": "npx",
      "Arguments": [
        "-y",
        "@teamshift/sandbox-mcp"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add teamshift-io-sandbox-mcp -t stdio -c npx -a -y @teamshift/sandbox-mcp
```

### Other

```json
{
  "mcpServers": {
    "teamshift-io-sandbox-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@teamshift/sandbox-mcp"
      ]
    }
  }
}
```

## 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.

### 2026-10-21 (score 79, +1)

No change was recorded against any check on this day. Stability & Change Management went from 53 to 57. That category is still filling its 30-day observation window: 16 days of observed history at the previous scan, 17 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-19 (score 78, +1)

No change was recorded against any check on this day. Stability & Change Management went from 47 to 50. That category is still filling its 30-day observation window: 14 days of observed history at the previous scan, 15 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-17 (score 77, +1)

No change was recorded against any check on this day. Stability & Change Management went from 40 to 43. That category is still filling its 30-day observation window: 12 days of observed history at the previous scan, 13 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-15 (score 76, +1)

No change was recorded against any check on this day. Stability & Change Management went from 33 to 37. That category is still filling its 30-day observation window: 10 days of observed history at the previous scan, 11 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-14 (score 75, +4)

- [functional improvement] Stability: unverified → 0.33

### 2026-10-12 (score 71, +1)

- [functional improvement] Security disclosure: unverified → pass

### 2026-10-11 (score 70, −1)

- [functional regression] Security disclosure: pass → unverified

### 2026-10-04 (score 71)

First indexed and scored.

## MCP tools (46)

### `crm_search_contacts` (~102 tokens)

Search contacts

Find people by name, email, phone digits or title. Use before any contact-specific action to get the contact id. Set missingInfo to find contacts lacking a valid email or phone.

Input parameters:

- `customerId` (string)
- `limit` (integer): Max rows to return (default 25, max 100).
- `missingInfo` (boolean)
- `offset` (integer): Rows to skip (default 0).
- `query` (string)

### `crm_get_contact` (~49 tokens)

Get contact

Full profile of one contact plus their customer account, leads, deals and recent calls. Use when you need context before replying to or changing a contact.

Input parameters:

- `contactId` (string, required): Contact id

### `crm_create_contact` (~77 tokens)

Create contact

Add a new person, optionally linked to a customer account. Search first to avoid creating duplicates.

Input parameters:

- `customerId` (string|null)
- `email` (string|null)
- `firstName` (string, required)
- `lastName` (string, required)
- `phone` (string|null)
- `title` (string|null)

### `crm_update_contact` (~105 tokens)

Update contact

Change a contact's name, email, phone, title or customer link. Omitted fields stay as they are; null clears a field. Use to fix missing or malformed contact info.

Input parameters:

- `contactId` (string, required): Contact id
- `customerId` (string|null)
- `email` (string|null)
- `firstName` (string)
- `lastName` (string)
- `phone` (string|null)
- `title` (string|null)

### `crm_merge_contacts` (~89 tokens)

Merge duplicate contacts

Fold a duplicate contact (mergeId) into the one to keep (keepId): empty fields are filled from the duplicate, all leads/deals/messages/calls are re-pointed, and the duplicate is deleted. Irreversible; confirm they are the same person first.

Input parameters:

- `keepId` (string, required): Contact to keep id
- `mergeId` (string, required): Duplicate contact to remove id

### `crm_list_customers` (~85 tokens)

List customers

List customer accounts (households or businesses), newest first. Filter by text, status or owner.

Input parameters:

- `limit` (integer): Max rows to return (default 25, max 100).
- `offset` (integer): Rows to skip (default 0).
- `ownerId` (string)
- `query` (string)
- `status` (string)

### `crm_get_customer` (~39 tokens)

Get customer

One customer account with its contacts, deals, jobs and invoices. Use for a full account picture.

Input parameters:

- `customerId` (string, required): Customer id

### `crm_list_employees` (~45 tokens)

List employees

Active staff with ids and roles. Use to pick an owner for a deal/lead or an assignee for a job or task.

Input parameters:

- `includeInactive` (boolean)

### `crm_list_leads` (~97 tokens)

List leads

Inbound leads, newest first. Set uncontacted: true to find leads nobody has responded to yet.

Input parameters:

- `contactId` (string)
- `limit` (integer): Max rows to return (default 25, max 100).
- `offset` (integer): Rows to skip (default 0).
- `ownerId` (string)
- `status` (string)
- `uncontacted` (boolean)

### `crm_update_lead` (~66 tokens)

Update lead

Change a lead's status or owner. Note: sending an email/SMS or logging an outbound call to the lead's contact marks it contacted automatically.

Input parameters:

- `leadId` (string, required): Lead id
- `ownerId` (string|null)
- `status` (string)

### `crm_list_deals` (~121 tokens)

List deals

Sales pipeline deals, most recently updated first. Use staleDays to find open deals with no activity for N days, openOnly to skip won/lost.

Input parameters:

- `contactId` (string)
- `customerId` (string)
- `limit` (integer): Max rows to return (default 25, max 100).
- `offset` (integer): Rows to skip (default 0).
- `openOnly` (boolean)
- `ownerId` (string)
- `stage` (string)
- `staleDays` (integer)

### `crm_get_deal` (~28 tokens)

Get deal

One deal with its quotes and contact.

Input parameters:

- `dealId` (string, required): Deal id

### `crm_update_deal` (~151 tokens)

Update deal

Move a deal to a new stage and/or set its next action, amount or owner. Stage moves must follow the pipeline (new→qualified→quote-sent→negotiation→won/lost; won is final; lost can reopen). Moving to lost requires lostReason. Set nextAction to {summary, dueOn} so every open deal has a next step.

Input parameters:

- `amountCents` (integer): Deal value in integer cents, e.g. 12500 = $125.00
- `dealId` (string, required): Deal id
- `lostReason` (string)
- `nextAction`
- `ownerId` (string|null)
- `stage` (string)

### `crm_list_quotes` (~98 tokens)

List quotes

Quotes/estimates. Set notFollowedUp: true to find sent quotes that never got a follow-up.

Input parameters:

- `customerId` (string)
- `dealId` (string)
- `limit` (integer): Max rows to return (default 25, max 100).
- `notFollowedUp` (boolean)
- `offset` (integer): Rows to skip (default 0).
- `status` (string)

### `crm_send_quote` (~49 tokens)

Send quote

Send a draft quote to the customer (status becomes sent). Only works on drafts.

Input parameters:

- `expiresOn` (string): Date, YYYY-MM-DD
- `quoteId` (string, required): Quote id

### `crm_follow_up_quote` (~102 tokens)

Follow up on quote

Message the deal's contact about a sent/viewed quote and record the follow-up on the quote. Write the full message body. Goes to the outbox; nothing is really delivered.

Input parameters:

- `body` (string, required)
- `channel` (string): Default email
- `contactId` (string): Defaults to the deal's contact
- `employeeId` (string): Sender; defaults to the company address
- `quoteId` (string, required): Quote id

### `crm_list_tasks` (~84 tokens)

List tasks

To-dos for staff. Filter by status, assignee or a related record id.

Input parameters:

- `assigneeId` (string)
- `limit` (integer): Max rows to return (default 25, max 100).
- `offset` (integer): Rows to skip (default 0).
- `relatedId` (string)
- `status` (string)

### `crm_create_task` (~70 tokens)

Create task

Create a to-do for a staff member, linked to the records it is about (relatedIds). Use when a human must do something you cannot.

Input parameters:

- `assigneeId` (string|null)
- `dueOn`
- `relatedIds` (array)
- `title` (string, required)

### `crm_complete_task` (~25 tokens)

Complete task

Mark an open task done.

Input parameters:

- `taskId` (string, required): Task id

### `inbox_list_threads` (~97 tokens)

List inbox threads

Email and SMS conversations, newest activity first, with unread counts and the last message snippet. Use unreadOnly to triage.

Input parameters:

- `channel` (string)
- `contactId` (string)
- `limit` (integer): Max rows to return (default 25, max 100).
- `offset` (integer): Rows to skip (default 0).
- `query` (string)
- `unreadOnly` (boolean)

### `inbox_get_thread` (~36 tokens)

Get thread

All messages in one thread, oldest first. Read before replying.

Input parameters:

- `threadId` (string, required): Thread (thr_…) id

### `inbox_send_email` (~115 tokens)

Send email

Send an email to a contact, raw address, or as a reply in an existing thread (threadId). Subject is required for new threads. Written to the outbox only; nothing leaves the sandbox.

Input parameters:

- `body` (string, required)
- `contactId` (string)
- `employeeId` (string): Sender; defaults to the company address
- `relatedIds` (array): Records this email is about
- `subject` (string)
- `threadId` (string)
- `to` (array)

### `inbox_send_sms` (~76 tokens)

Send SMS

Send a text message to a contact, raw number, or within an SMS thread. Outbox only.

Input parameters:

- `body` (string, required)
- `contactId` (string)
- `employeeId` (string)
- `relatedIds` (array)
- `threadId` (string)
- `to` (string)

### `inbox_mark_thread_read` (~32 tokens)

Mark thread read

Mark every message in a thread as read after handling it.

Input parameters:

- `threadId` (string, required): Thread id

### `calendar_list_jobs` (~127 tokens)

List jobs

Scheduled work (appointments / work orders / projects), latest start first. Filter by status, customer, assignee, or a start-time window [from, to).

Input parameters:

- `assigneeId` (string)
- `customerId` (string)
- `from` (string): Inclusive; YYYY-MM-DD or datetime
- `limit` (integer): Max rows to return (default 25, max 100).
- `offset` (integer): Rows to skip (default 0).
- `status` (string)
- `to` (string): Exclusive; YYYY-MM-DD or datetime

### `calendar_get_job` (~27 tokens)

Get job

One job with its customer and invoices.

Input parameters:

- `jobId` (string, required): Job id

### `calendar_schedule_job` (~148 tokens)

Schedule job

Book new work for a customer. Assignees must be active employees and not double-booked (pass allowOverlap to override).

Input parameters:

- `allowOverlap` (boolean)
- `assigneeIds` (array)
- `customerId` (string, required): Customer id
- `notes` (string|null)
- `quoteId` (string|null)
- `scheduledEnd` (string, required): ISO-8601 datetime, e.g. 2026-10-01T15:00:00Z
- `scheduledStart` (string, required): ISO-8601 datetime, e.g. 2026-10-01T15:00:00Z
- `title` (string, required)

### `calendar_reschedule_job` (~141 tokens)

Reschedule job

Move a scheduled or no-show job to a new time window (optionally new assignees). Does NOT notify the customer; send them a message separately if policy requires.

Input parameters:

- `allowOverlap` (boolean)
- `assigneeIds` (array)
- `jobId` (string, required): Job id
- `reason` (string)
- `scheduledEnd` (string, required): ISO-8601 datetime, e.g. 2026-10-01T15:00:00Z
- `scheduledStart` (string, required): ISO-8601 datetime, e.g. 2026-10-01T15:00:00Z

### `calendar_cancel_job` (~42 tokens)

Cancel job

Cancel a scheduled or in-progress job. Requires a reason. Cannot be undone.

Input parameters:

- `jobId` (string, required): Job id
- `reason` (string, required)

### `calendar_complete_job` (~40 tokens)

Complete job

Mark a scheduled or in-progress job completed now, with optional notes.

Input parameters:

- `jobId` (string, required): Job id
- `notes` (string)

### `invoicing_list_invoices` (~92 tokens)

List invoices

Invoices, newest first. Set overdue: true for unpaid invoices past their due date.

Input parameters:

- `customerId` (string)
- `jobId` (string)
- `limit` (integer): Max rows to return (default 25, max 100).
- `offset` (integer): Rows to skip (default 0).
- `overdue` (boolean)
- `status` (string)

### `invoicing_get_invoice` (~29 tokens)

Get invoice

One invoice with its payments and balance due.

Input parameters:

- `invoiceId` (string, required): Invoice id

### `invoicing_create_invoice` (~94 tokens)

Create invoice

Issue an invoice to a customer (optionally for a job of that customer). Total is computed from line items; due date defaults to company payment terms.

Input parameters:

- `customerId` (string, required): Customer id
- `dueOn` (string): Date, YYYY-MM-DD
- `issuedOn` (string): Date, YYYY-MM-DD
- `jobId` (string|null)
- `lineItems` (array, required)

### `invoicing_send_reminder` (~82 tokens)

Send invoice reminder

Send a payment reminder for an open or partially-paid invoice to the customer's contact (outbox only). A default polite body is used if none is given.

Input parameters:

- `body` (string)
- `channel` (string)
- `contactId` (string)
- `employeeId` (string)
- `invoiceId` (string, required): Invoice id

### `invoicing_void_invoice` (~53 tokens)

Void invoice

Void an invoice that has no payments applied (e.g. issued against the wrong job). Irreversible; requires a reason.

Input parameters:

- `invoiceId` (string, required): Invoice id
- `reason` (string, required)

### `invoicing_record_payment` (~81 tokens)

Record payment

Record money received against an invoice. Cannot exceed the balance; void, draft and paid invoices are rejected.

Input parameters:

- `amountCents` (integer, required): Amount received in integer cents, e.g. 12500 = $125.00
- `invoiceId` (string, required): Invoice id
- `method` (string, required)
- `reference` (string)

### `invoicing_list_payments` (~86 tokens)

List payments

Payments received, newest first. Set unmatched: true for payments not yet applied to any invoice.

Input parameters:

- `customerId` (string)
- `invoiceId` (string)
- `limit` (integer): Max rows to return (default 25, max 100).
- `offset` (integer): Rows to skip (default 0).
- `unmatched` (boolean)

### `invoicing_match_payment` (~51 tokens)

Match payment to invoice

Apply an unmatched payment to the invoice it pays. Customer must match and amount must not exceed the balance.

Input parameters:

- `invoiceId` (string, required): Invoice id
- `paymentId` (string, required): Payment id

### `phone_list_calls` (~102 tokens)

List calls

Phone call log, newest first. Use outcome: missed and direction: inbound to find calls that may need a callback.

Input parameters:

- `contactId` (string)
- `direction` (string)
- `limit` (integer): Max rows to return (default 25, max 100).
- `offset` (integer): Rows to skip (default 0).
- `outcome` (string)
- `since` (string): Inclusive; YYYY-MM-DD or datetime

### `phone_log_call` (~97 tokens)

Log call

Record a call you made or took (e.g. a callback), with outcome and a short summary. No real call is placed.

Input parameters:

- `contactId` (string)
- `direction` (string): Default outbound
- `durationSec` (integer)
- `employeeId` (string)
- `outcome` (string, required)
- `phone` (string): Defaults to the contact's phone
- `summary` (string|null)

### `admin_get_company_policies` (~32 tokens)

Get company policies

The company's plain-language operating rules (response times, follow-up cadence, payment terms...). Read before acting.

### `admin_get_company` (~23 tokens)

Get company profile

Company profile plus the simulated current time and today's date in company time.

### `admin_get_audit_log` (~47 tokens)

Get audit log

Every mutation attempted so far (tool, input, ok/error, changed ids), in order. Use sinceSeq to fetch only newer entries.

Input parameters:

- `sinceSeq` (integer)

### `admin_reset` (~30 tokens)

Reset sandbox

Restore the original dataset and clock, discarding every change. The audit log is kept and records the reset.

### `admin_save_state` (~28 tokens)

Save state

Write the current dataset, audit log and outbox to the --state-out file for grading.

### `admin_advance_clock` (~35 tokens)

Advance clock

Move simulated time forward by N minutes (e.g. to simulate waiting for a reply).

Input parameters:

- `minutes` (integer, required)

## Diagnostics

Captured diagnostic sections: Provenance, Dependencies. The full working is on the page: https://verifymcp.io/servers/teamshift-io-sandbox-mcp/teamshift-sandbox-mcp#diagnostics

## Score history

- 2026-10-23: 79
- 2026-10-22: 79
- 2026-10-21: 79
- 2026-10-20: 78
- 2026-10-19: 78
- 2026-10-18: 77
- 2026-10-17: 77
- 2026-10-16: 76
- 2026-10-15: 76
- 2026-10-14: 75
- 2026-10-13: 71
- 2026-10-12: 71
- 2026-10-11: 70
- 2026-10-04: 71

## Common questions

### What is the SMB Sandbox MCP server?

SMB Sandbox is an MCP server listed in the public MCP registry as io.github.teamshift-io/sandbox-mcp. Mock small-business CRM, inbox, calendar, invoicing and phone tools for testing AI agents safely. This page covers its npm package (@teamshift/sandbox-mcp).

### Is the SMB Sandbox MCP server safe to use?

SMB Sandbox scores 79 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 23 October 2026. It declares no install or post-install scripts. 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 SMB Sandbox MCP server expose?

SMB Sandbox exposes 46 tools: crm_search_contacts, crm_get_contact, crm_create_contact, crm_update_contact, crm_merge_contacts, and 41 more. Their descriptions and schemas cost roughly 3,325 tokens of context every time the server is loaded.

### Is the SMB Sandbox MCP server still maintained?

SMB Sandbox is still listed as active in the MCP registry. We last reached this channel on 23 October 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What licence is the SMB Sandbox MCP server under?

SMB Sandbox declares the Apache-2.0 licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

## Links

- npm package: https://www.npmjs.com/package/@teamshift/sandbox-mcp
- Socket report: https://socket.dev/npm/package/@teamshift/sandbox-mcp
- Repository: https://github.com/teamshift-io/smb-sandbox
- Website: https://teamshift.io/open-source/smb-sandbox
- Changelog RSS feed: https://verifymcp.io/servers/teamshift-io-sandbox-mcp/teamshift-sandbox-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/teamshift-io-sandbox-mcp/teamshift-sandbox-mcp.json
- HTML version of this page: https://verifymcp.io/servers/teamshift-io-sandbox-mcp/teamshift-sandbox-mcp